diff --git a/CHANGELOG.md b/CHANGELOG.md index 98ba0c4c2..f7416361d 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -5,6 +5,9 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). +### Added +- **🌐 Enhanced Translations**: Added Slovak language, improved Czech language. + ## [0.4.8] - 2024-12-07 ### Added diff --git a/backend/open_webui/apps/audio/main.py b/backend/open_webui/apps/audio/main.py deleted file mode 100644 index a3972f19f..000000000 --- a/backend/open_webui/apps/audio/main.py +++ /dev/null @@ -1,703 +0,0 @@ -import hashlib -import json -import logging -import os -import uuid -from functools import lru_cache -from pathlib import Path -from pydub import AudioSegment -from pydub.silence import split_on_silence - -import aiohttp -import aiofiles -import requests -from open_webui.config import ( - AUDIO_STT_ENGINE, - AUDIO_STT_MODEL, - AUDIO_STT_OPENAI_API_BASE_URL, - AUDIO_STT_OPENAI_API_KEY, - AUDIO_TTS_API_KEY, - AUDIO_TTS_ENGINE, - AUDIO_TTS_MODEL, - AUDIO_TTS_OPENAI_API_BASE_URL, - AUDIO_TTS_OPENAI_API_KEY, - AUDIO_TTS_SPLIT_ON, - AUDIO_TTS_VOICE, - AUDIO_TTS_AZURE_SPEECH_REGION, - AUDIO_TTS_AZURE_SPEECH_OUTPUT_FORMAT, - CACHE_DIR, - CORS_ALLOW_ORIGIN, - WHISPER_MODEL, - WHISPER_MODEL_AUTO_UPDATE, - WHISPER_MODEL_DIR, - AppConfig, -) - -from open_webui.constants import ERROR_MESSAGES -from open_webui.env import ( - ENV, - SRC_LOG_LEVELS, - DEVICE_TYPE, - ENABLE_FORWARD_USER_INFO_HEADERS, -) - -from fastapi import Depends, FastAPI, File, HTTPException, Request, UploadFile, status -from fastapi.middleware.cors import CORSMiddleware -from fastapi.responses import FileResponse -from pydantic import BaseModel -from open_webui.utils.auth import get_admin_user, get_verified_user - -# Constants -MAX_FILE_SIZE_MB = 25 -MAX_FILE_SIZE = MAX_FILE_SIZE_MB * 1024 * 1024 # Convert MB to bytes - - -log = logging.getLogger(__name__) -log.setLevel(SRC_LOG_LEVELS["AUDIO"]) - -app = FastAPI( - docs_url="/docs" if ENV == "dev" else None, - openapi_url="/openapi.json" if ENV == "dev" else None, - redoc_url=None, -) - -app.add_middleware( - CORSMiddleware, - allow_origins=CORS_ALLOW_ORIGIN, - allow_credentials=True, - allow_methods=["*"], - allow_headers=["*"], -) - -app.state.config = AppConfig() - -app.state.config.STT_OPENAI_API_BASE_URL = AUDIO_STT_OPENAI_API_BASE_URL -app.state.config.STT_OPENAI_API_KEY = AUDIO_STT_OPENAI_API_KEY -app.state.config.STT_ENGINE = AUDIO_STT_ENGINE -app.state.config.STT_MODEL = AUDIO_STT_MODEL - -app.state.config.WHISPER_MODEL = WHISPER_MODEL -app.state.faster_whisper_model = None - -app.state.config.TTS_OPENAI_API_BASE_URL = AUDIO_TTS_OPENAI_API_BASE_URL -app.state.config.TTS_OPENAI_API_KEY = AUDIO_TTS_OPENAI_API_KEY -app.state.config.TTS_ENGINE = AUDIO_TTS_ENGINE -app.state.config.TTS_MODEL = AUDIO_TTS_MODEL -app.state.config.TTS_VOICE = AUDIO_TTS_VOICE -app.state.config.TTS_API_KEY = AUDIO_TTS_API_KEY -app.state.config.TTS_SPLIT_ON = AUDIO_TTS_SPLIT_ON - - -app.state.speech_synthesiser = None -app.state.speech_speaker_embeddings_dataset = None - -app.state.config.TTS_AZURE_SPEECH_REGION = AUDIO_TTS_AZURE_SPEECH_REGION -app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT = AUDIO_TTS_AZURE_SPEECH_OUTPUT_FORMAT - -# setting device type for whisper model -whisper_device_type = DEVICE_TYPE if DEVICE_TYPE and DEVICE_TYPE == "cuda" else "cpu" -log.info(f"whisper_device_type: {whisper_device_type}") - -SPEECH_CACHE_DIR = Path(CACHE_DIR).joinpath("./audio/speech/") -SPEECH_CACHE_DIR.mkdir(parents=True, exist_ok=True) - - -def set_faster_whisper_model(model: str, auto_update: bool = False): - if model and app.state.config.STT_ENGINE == "": - from faster_whisper import WhisperModel - - faster_whisper_kwargs = { - "model_size_or_path": model, - "device": whisper_device_type, - "compute_type": "int8", - "download_root": WHISPER_MODEL_DIR, - "local_files_only": not auto_update, - } - - try: - app.state.faster_whisper_model = WhisperModel(**faster_whisper_kwargs) - except Exception: - log.warning( - "WhisperModel initialization failed, attempting download with local_files_only=False" - ) - faster_whisper_kwargs["local_files_only"] = False - app.state.faster_whisper_model = WhisperModel(**faster_whisper_kwargs) - - else: - app.state.faster_whisper_model = None - - -class TTSConfigForm(BaseModel): - OPENAI_API_BASE_URL: str - OPENAI_API_KEY: str - API_KEY: str - ENGINE: str - MODEL: str - VOICE: str - SPLIT_ON: str - AZURE_SPEECH_REGION: str - AZURE_SPEECH_OUTPUT_FORMAT: str - - -class STTConfigForm(BaseModel): - OPENAI_API_BASE_URL: str - OPENAI_API_KEY: str - ENGINE: str - MODEL: str - WHISPER_MODEL: str - - -class AudioConfigUpdateForm(BaseModel): - tts: TTSConfigForm - stt: STTConfigForm - - -from pydub import AudioSegment -from pydub.utils import mediainfo - - -def is_mp4_audio(file_path): - """Check if the given file is an MP4 audio file.""" - if not os.path.isfile(file_path): - print(f"File not found: {file_path}") - return False - - info = mediainfo(file_path) - if ( - info.get("codec_name") == "aac" - and info.get("codec_type") == "audio" - and info.get("codec_tag_string") == "mp4a" - ): - return True - return False - - -def convert_mp4_to_wav(file_path, output_path): - """Convert MP4 audio file to WAV format.""" - audio = AudioSegment.from_file(file_path, format="mp4") - audio.export(output_path, format="wav") - print(f"Converted {file_path} to {output_path}") - - -@app.get("/config") -async def get_audio_config(user=Depends(get_admin_user)): - return { - "tts": { - "OPENAI_API_BASE_URL": app.state.config.TTS_OPENAI_API_BASE_URL, - "OPENAI_API_KEY": app.state.config.TTS_OPENAI_API_KEY, - "API_KEY": app.state.config.TTS_API_KEY, - "ENGINE": app.state.config.TTS_ENGINE, - "MODEL": app.state.config.TTS_MODEL, - "VOICE": app.state.config.TTS_VOICE, - "SPLIT_ON": app.state.config.TTS_SPLIT_ON, - "AZURE_SPEECH_REGION": app.state.config.TTS_AZURE_SPEECH_REGION, - "AZURE_SPEECH_OUTPUT_FORMAT": app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT, - }, - "stt": { - "OPENAI_API_BASE_URL": app.state.config.STT_OPENAI_API_BASE_URL, - "OPENAI_API_KEY": app.state.config.STT_OPENAI_API_KEY, - "ENGINE": app.state.config.STT_ENGINE, - "MODEL": app.state.config.STT_MODEL, - "WHISPER_MODEL": app.state.config.WHISPER_MODEL, - }, - } - - -@app.post("/config/update") -async def update_audio_config( - form_data: AudioConfigUpdateForm, user=Depends(get_admin_user) -): - app.state.config.TTS_OPENAI_API_BASE_URL = form_data.tts.OPENAI_API_BASE_URL - app.state.config.TTS_OPENAI_API_KEY = form_data.tts.OPENAI_API_KEY - app.state.config.TTS_API_KEY = form_data.tts.API_KEY - app.state.config.TTS_ENGINE = form_data.tts.ENGINE - app.state.config.TTS_MODEL = form_data.tts.MODEL - app.state.config.TTS_VOICE = form_data.tts.VOICE - app.state.config.TTS_SPLIT_ON = form_data.tts.SPLIT_ON - app.state.config.TTS_AZURE_SPEECH_REGION = form_data.tts.AZURE_SPEECH_REGION - app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT = ( - form_data.tts.AZURE_SPEECH_OUTPUT_FORMAT - ) - - app.state.config.STT_OPENAI_API_BASE_URL = form_data.stt.OPENAI_API_BASE_URL - app.state.config.STT_OPENAI_API_KEY = form_data.stt.OPENAI_API_KEY - app.state.config.STT_ENGINE = form_data.stt.ENGINE - app.state.config.STT_MODEL = form_data.stt.MODEL - app.state.config.WHISPER_MODEL = form_data.stt.WHISPER_MODEL - set_faster_whisper_model(form_data.stt.WHISPER_MODEL, WHISPER_MODEL_AUTO_UPDATE) - - return { - "tts": { - "OPENAI_API_BASE_URL": app.state.config.TTS_OPENAI_API_BASE_URL, - "OPENAI_API_KEY": app.state.config.TTS_OPENAI_API_KEY, - "API_KEY": app.state.config.TTS_API_KEY, - "ENGINE": app.state.config.TTS_ENGINE, - "MODEL": app.state.config.TTS_MODEL, - "VOICE": app.state.config.TTS_VOICE, - "SPLIT_ON": app.state.config.TTS_SPLIT_ON, - "AZURE_SPEECH_REGION": app.state.config.TTS_AZURE_SPEECH_REGION, - "AZURE_SPEECH_OUTPUT_FORMAT": app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT, - }, - "stt": { - "OPENAI_API_BASE_URL": app.state.config.STT_OPENAI_API_BASE_URL, - "OPENAI_API_KEY": app.state.config.STT_OPENAI_API_KEY, - "ENGINE": app.state.config.STT_ENGINE, - "MODEL": app.state.config.STT_MODEL, - "WHISPER_MODEL": app.state.config.WHISPER_MODEL, - }, - } - - -def load_speech_pipeline(): - from transformers import pipeline - from datasets import load_dataset - - if app.state.speech_synthesiser is None: - app.state.speech_synthesiser = pipeline( - "text-to-speech", "microsoft/speecht5_tts" - ) - - if app.state.speech_speaker_embeddings_dataset is None: - app.state.speech_speaker_embeddings_dataset = load_dataset( - "Matthijs/cmu-arctic-xvectors", split="validation" - ) - - -@app.post("/speech") -async def speech(request: Request, user=Depends(get_verified_user)): - body = await request.body() - name = hashlib.sha256(body).hexdigest() - - file_path = SPEECH_CACHE_DIR.joinpath(f"{name}.mp3") - file_body_path = SPEECH_CACHE_DIR.joinpath(f"{name}.json") - - # Check if the file already exists in the cache - if file_path.is_file(): - return FileResponse(file_path) - - if app.state.config.TTS_ENGINE == "openai": - headers = {} - headers["Authorization"] = f"Bearer {app.state.config.TTS_OPENAI_API_KEY}" - headers["Content-Type"] = "application/json" - - if ENABLE_FORWARD_USER_INFO_HEADERS: - headers["X-OpenWebUI-User-Name"] = user.name - headers["X-OpenWebUI-User-Id"] = user.id - headers["X-OpenWebUI-User-Email"] = user.email - headers["X-OpenWebUI-User-Role"] = user.role - - try: - body = body.decode("utf-8") - body = json.loads(body) - body["model"] = app.state.config.TTS_MODEL - body = json.dumps(body).encode("utf-8") - except Exception: - pass - - try: - async with aiohttp.ClientSession() as session: - async with session.post( - url=f"{app.state.config.TTS_OPENAI_API_BASE_URL}/audio/speech", - data=body, - headers=headers, - ) as r: - r.raise_for_status() - async with aiofiles.open(file_path, "wb") as f: - await f.write(await r.read()) - - async with aiofiles.open(file_body_path, "w") as f: - await f.write(json.dumps(json.loads(body.decode("utf-8")))) - - return FileResponse(file_path) - - except Exception as e: - log.exception(e) - error_detail = "Open WebUI: Server Connection Error" - try: - if r.status != 200: - res = await r.json() - if "error" in res: - error_detail = f"External: {res['error']['message']}" - except Exception: - error_detail = f"External: {e}" - - raise HTTPException( - status_code=getattr(r, "status", 500), - detail=error_detail, - ) - - elif app.state.config.TTS_ENGINE == "elevenlabs": - try: - payload = json.loads(body.decode("utf-8")) - except Exception as e: - log.exception(e) - raise HTTPException(status_code=400, detail="Invalid JSON payload") - - voice_id = payload.get("voice", "") - if voice_id not in get_available_voices(): - raise HTTPException( - status_code=400, - detail="Invalid voice id", - ) - - url = f"https://api.elevenlabs.io/v1/text-to-speech/{voice_id}" - headers = { - "Accept": "audio/mpeg", - "Content-Type": "application/json", - "xi-api-key": app.state.config.TTS_API_KEY, - } - data = { - "text": payload["input"], - "model_id": app.state.config.TTS_MODEL, - "voice_settings": {"stability": 0.5, "similarity_boost": 0.5}, - } - - try: - async with aiohttp.ClientSession() as session: - async with session.post(url, json=data, headers=headers) as r: - r.raise_for_status() - async with aiofiles.open(file_path, "wb") as f: - await f.write(await r.read()) - - async with aiofiles.open(file_body_path, "w") as f: - await f.write(json.dumps(json.loads(body.decode("utf-8")))) - - return FileResponse(file_path) - - except Exception as e: - log.exception(e) - error_detail = "Open WebUI: Server Connection Error" - try: - if r.status != 200: - res = await r.json() - if "error" in res: - error_detail = f"External: {res['error']['message']}" - except Exception: - error_detail = f"External: {e}" - - raise HTTPException( - status_code=getattr(r, "status", 500), - detail=error_detail, - ) - - elif app.state.config.TTS_ENGINE == "azure": - try: - payload = json.loads(body.decode("utf-8")) - except Exception as e: - log.exception(e) - raise HTTPException(status_code=400, detail="Invalid JSON payload") - - region = app.state.config.TTS_AZURE_SPEECH_REGION - language = app.state.config.TTS_VOICE - locale = "-".join(app.state.config.TTS_VOICE.split("-")[:1]) - output_format = app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT - url = f"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1" - - headers = { - "Ocp-Apim-Subscription-Key": app.state.config.TTS_API_KEY, - "Content-Type": "application/ssml+xml", - "X-Microsoft-OutputFormat": output_format, - } - - data = f""" - {payload["input"]} - """ - - try: - async with aiohttp.ClientSession() as session: - async with session.post(url, headers=headers, data=data) as response: - if response.status == 200: - async with aiofiles.open(file_path, "wb") as f: - await f.write(await response.read()) - return FileResponse(file_path) - else: - error_msg = f"Error synthesizing speech - {response.reason}" - log.error(error_msg) - raise HTTPException(status_code=500, detail=error_msg) - except Exception as e: - log.exception(e) - raise HTTPException(status_code=500, detail=str(e)) - elif app.state.config.TTS_ENGINE == "transformers": - payload = None - try: - payload = json.loads(body.decode("utf-8")) - except Exception as e: - log.exception(e) - raise HTTPException(status_code=400, detail="Invalid JSON payload") - - import torch - import soundfile as sf - - load_speech_pipeline() - - embeddings_dataset = app.state.speech_speaker_embeddings_dataset - - speaker_index = 6799 - try: - speaker_index = embeddings_dataset["filename"].index( - app.state.config.TTS_MODEL - ) - except Exception: - pass - - speaker_embedding = torch.tensor( - embeddings_dataset[speaker_index]["xvector"] - ).unsqueeze(0) - - speech = app.state.speech_synthesiser( - payload["input"], - forward_params={"speaker_embeddings": speaker_embedding}, - ) - - sf.write(file_path, speech["audio"], samplerate=speech["sampling_rate"]) - with open(file_body_path, "w") as f: - json.dump(json.loads(body.decode("utf-8")), f) - - return FileResponse(file_path) - - -def transcribe(file_path): - print("transcribe", file_path) - filename = os.path.basename(file_path) - file_dir = os.path.dirname(file_path) - id = filename.split(".")[0] - - if app.state.config.STT_ENGINE == "": - if app.state.faster_whisper_model is None: - set_faster_whisper_model(app.state.config.WHISPER_MODEL) - - model = app.state.faster_whisper_model - segments, info = model.transcribe(file_path, beam_size=5) - log.info( - "Detected language '%s' with probability %f" - % (info.language, info.language_probability) - ) - - transcript = "".join([segment.text for segment in list(segments)]) - data = {"text": transcript.strip()} - - # save the transcript to a json file - transcript_file = f"{file_dir}/{id}.json" - with open(transcript_file, "w") as f: - json.dump(data, f) - - log.debug(data) - return data - elif app.state.config.STT_ENGINE == "openai": - if is_mp4_audio(file_path): - print("is_mp4_audio") - os.rename(file_path, file_path.replace(".wav", ".mp4")) - # Convert MP4 audio file to WAV format - convert_mp4_to_wav(file_path.replace(".wav", ".mp4"), file_path) - - headers = {"Authorization": f"Bearer {app.state.config.STT_OPENAI_API_KEY}"} - - files = {"file": (filename, open(file_path, "rb"))} - data = {"model": app.state.config.STT_MODEL} - - log.debug(files, data) - - r = None - try: - r = requests.post( - url=f"{app.state.config.STT_OPENAI_API_BASE_URL}/audio/transcriptions", - headers=headers, - files=files, - data=data, - ) - - r.raise_for_status() - - data = r.json() - - # save the transcript to a json file - transcript_file = f"{file_dir}/{id}.json" - with open(transcript_file, "w") as f: - json.dump(data, f) - - print(data) - return data - except Exception as e: - log.exception(e) - error_detail = "Open WebUI: Server Connection Error" - if r is not None: - try: - res = r.json() - if "error" in res: - error_detail = f"External: {res['error']['message']}" - except Exception: - error_detail = f"External: {e}" - - raise Exception(error_detail) - - -@app.post("/transcriptions") -def transcription( - file: UploadFile = File(...), - user=Depends(get_verified_user), -): - log.info(f"file.content_type: {file.content_type}") - - if file.content_type not in ["audio/mpeg", "audio/wav", "audio/ogg", "audio/x-m4a"]: - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=ERROR_MESSAGES.FILE_NOT_SUPPORTED, - ) - - try: - ext = file.filename.split(".")[-1] - id = uuid.uuid4() - - filename = f"{id}.{ext}" - contents = file.file.read() - - file_dir = f"{CACHE_DIR}/audio/transcriptions" - os.makedirs(file_dir, exist_ok=True) - file_path = f"{file_dir}/{filename}" - - with open(file_path, "wb") as f: - f.write(contents) - - try: - if os.path.getsize(file_path) > MAX_FILE_SIZE: # file is bigger than 25MB - log.debug(f"File size is larger than {MAX_FILE_SIZE_MB}MB") - audio = AudioSegment.from_file(file_path) - audio = audio.set_frame_rate(16000).set_channels(1) # Compress audio - compressed_path = f"{file_dir}/{id}_compressed.opus" - audio.export(compressed_path, format="opus", bitrate="32k") - log.debug(f"Compressed audio to {compressed_path}") - file_path = compressed_path - - if ( - os.path.getsize(file_path) > MAX_FILE_SIZE - ): # Still larger than 25MB after compression - log.debug( - f"Compressed file size is still larger than {MAX_FILE_SIZE_MB}MB: {os.path.getsize(file_path)}" - ) - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=ERROR_MESSAGES.FILE_TOO_LARGE( - size=f"{MAX_FILE_SIZE_MB}MB" - ), - ) - - data = transcribe(file_path) - else: - data = transcribe(file_path) - - file_path = file_path.split("/")[-1] - return {**data, "filename": file_path} - except Exception as e: - log.exception(e) - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=ERROR_MESSAGES.DEFAULT(e), - ) - - except Exception as e: - log.exception(e) - - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=ERROR_MESSAGES.DEFAULT(e), - ) - - -def get_available_models() -> list[dict]: - if app.state.config.TTS_ENGINE == "openai": - return [{"id": "tts-1"}, {"id": "tts-1-hd"}] - elif app.state.config.TTS_ENGINE == "elevenlabs": - headers = { - "xi-api-key": app.state.config.TTS_API_KEY, - "Content-Type": "application/json", - } - - try: - response = requests.get( - "https://api.elevenlabs.io/v1/models", headers=headers, timeout=5 - ) - response.raise_for_status() - models = response.json() - return [ - {"name": model["name"], "id": model["model_id"]} for model in models - ] - except requests.RequestException as e: - log.error(f"Error fetching voices: {str(e)}") - return [] - - -@app.get("/models") -async def get_models(user=Depends(get_verified_user)): - return {"models": get_available_models()} - - -def get_available_voices() -> dict: - """Returns {voice_id: voice_name} dict""" - ret = {} - if app.state.config.TTS_ENGINE == "openai": - ret = { - "alloy": "alloy", - "echo": "echo", - "fable": "fable", - "onyx": "onyx", - "nova": "nova", - "shimmer": "shimmer", - } - elif app.state.config.TTS_ENGINE == "elevenlabs": - try: - ret = get_elevenlabs_voices() - except Exception: - # Avoided @lru_cache with exception - pass - elif app.state.config.TTS_ENGINE == "azure": - try: - region = app.state.config.TTS_AZURE_SPEECH_REGION - url = f"https://{region}.tts.speech.microsoft.com/cognitiveservices/voices/list" - headers = {"Ocp-Apim-Subscription-Key": app.state.config.TTS_API_KEY} - - response = requests.get(url, headers=headers) - response.raise_for_status() - voices = response.json() - for voice in voices: - ret[voice["ShortName"]] = ( - f"{voice['DisplayName']} ({voice['ShortName']})" - ) - except requests.RequestException as e: - log.error(f"Error fetching voices: {str(e)}") - - return ret - - -@lru_cache -def get_elevenlabs_voices() -> dict: - """ - Note, set the following in your .env file to use Elevenlabs: - AUDIO_TTS_ENGINE=elevenlabs - AUDIO_TTS_API_KEY=sk_... # Your Elevenlabs API key - AUDIO_TTS_VOICE=EXAVITQu4vr4xnSDxMaL # From https://api.elevenlabs.io/v1/voices - AUDIO_TTS_MODEL=eleven_multilingual_v2 - """ - headers = { - "xi-api-key": app.state.config.TTS_API_KEY, - "Content-Type": "application/json", - } - try: - # TODO: Add retries - response = requests.get("https://api.elevenlabs.io/v1/voices", headers=headers) - response.raise_for_status() - voices_data = response.json() - - voices = {} - for voice in voices_data.get("voices", []): - voices[voice["voice_id"]] = voice["name"] - except requests.RequestException as e: - # Avoid @lru_cache with exception - log.error(f"Error fetching voices: {str(e)}") - raise RuntimeError(f"Error fetching voices: {str(e)}") - - return voices - - -@app.get("/voices") -async def get_voices(user=Depends(get_verified_user)): - return {"voices": [{"id": k, "name": v} for k, v in get_available_voices().items()]} diff --git a/backend/open_webui/apps/retrieval/vector/connector.py b/backend/open_webui/apps/retrieval/vector/connector.py deleted file mode 100644 index 528835b56..000000000 --- a/backend/open_webui/apps/retrieval/vector/connector.py +++ /dev/null @@ -1,22 +0,0 @@ -from open_webui.config import VECTOR_DB - -if VECTOR_DB == "milvus": - from open_webui.apps.retrieval.vector.dbs.milvus import MilvusClient - - VECTOR_DB_CLIENT = MilvusClient() -elif VECTOR_DB == "qdrant": - from open_webui.apps.retrieval.vector.dbs.qdrant import QdrantClient - - VECTOR_DB_CLIENT = QdrantClient() -elif VECTOR_DB == "opensearch": - from open_webui.apps.retrieval.vector.dbs.opensearch import OpenSearchClient - - VECTOR_DB_CLIENT = OpenSearchClient() -elif VECTOR_DB == "pgvector": - from open_webui.apps.retrieval.vector.dbs.pgvector import PgvectorClient - - VECTOR_DB_CLIENT = PgvectorClient() -else: - from open_webui.apps.retrieval.vector.dbs.chroma import ChromaClient - - VECTOR_DB_CLIENT = ChromaClient() diff --git a/backend/open_webui/apps/webui/main.py b/backend/open_webui/apps/webui/main.py deleted file mode 100644 index 054c6280e..000000000 --- a/backend/open_webui/apps/webui/main.py +++ /dev/null @@ -1,506 +0,0 @@ -import inspect -import json -import logging -import time -from typing import AsyncGenerator, Generator, Iterator - -from open_webui.apps.socket.main import get_event_call, get_event_emitter -from open_webui.apps.webui.models.functions import Functions -from open_webui.apps.webui.models.models import Models -from open_webui.apps.webui.routers import ( - auths, - chats, - folders, - configs, - groups, - files, - functions, - memories, - models, - knowledge, - prompts, - evaluations, - tools, - users, - utils, -) -from open_webui.apps.webui.utils import load_function_module_by_id -from open_webui.config import ( - ADMIN_EMAIL, - CORS_ALLOW_ORIGIN, - DEFAULT_MODELS, - DEFAULT_PROMPT_SUGGESTIONS, - DEFAULT_USER_ROLE, - MODEL_ORDER_LIST, - ENABLE_COMMUNITY_SHARING, - ENABLE_LOGIN_FORM, - ENABLE_MESSAGE_RATING, - ENABLE_SIGNUP, - ENABLE_API_KEY, - ENABLE_EVALUATION_ARENA_MODELS, - EVALUATION_ARENA_MODELS, - DEFAULT_ARENA_MODEL, - JWT_EXPIRES_IN, - ENABLE_OAUTH_ROLE_MANAGEMENT, - OAUTH_ROLES_CLAIM, - OAUTH_EMAIL_CLAIM, - OAUTH_PICTURE_CLAIM, - OAUTH_USERNAME_CLAIM, - OAUTH_ALLOWED_ROLES, - OAUTH_ADMIN_ROLES, - SHOW_ADMIN_DETAILS, - USER_PERMISSIONS, - WEBHOOK_URL, - WEBUI_AUTH, - WEBUI_BANNERS, - ENABLE_LDAP, - LDAP_SERVER_LABEL, - LDAP_SERVER_HOST, - LDAP_SERVER_PORT, - LDAP_ATTRIBUTE_FOR_USERNAME, - LDAP_SEARCH_FILTERS, - LDAP_SEARCH_BASE, - LDAP_APP_DN, - LDAP_APP_PASSWORD, - LDAP_USE_TLS, - LDAP_CA_CERT_FILE, - LDAP_CIPHERS, - AppConfig, -) -from open_webui.env import ( - ENV, - SRC_LOG_LEVELS, - WEBUI_AUTH_TRUSTED_EMAIL_HEADER, - WEBUI_AUTH_TRUSTED_NAME_HEADER, -) -from fastapi import FastAPI -from fastapi.middleware.cors import CORSMiddleware -from fastapi.responses import StreamingResponse -from pydantic import BaseModel -from open_webui.utils.misc import ( - openai_chat_chunk_message_template, - openai_chat_completion_message_template, -) -from open_webui.utils.payload import ( - apply_model_params_to_body_openai, - apply_model_system_prompt_to_body, -) - - -from open_webui.utils.tools import get_tools - -app = FastAPI( - docs_url="/docs" if ENV == "dev" else None, - openapi_url="/openapi.json" if ENV == "dev" else None, - redoc_url=None, -) - -log = logging.getLogger(__name__) -log.setLevel(SRC_LOG_LEVELS["MAIN"]) - -app.state.config = AppConfig() - -app.state.config.ENABLE_SIGNUP = ENABLE_SIGNUP -app.state.config.ENABLE_LOGIN_FORM = ENABLE_LOGIN_FORM -app.state.config.ENABLE_API_KEY = ENABLE_API_KEY - -app.state.config.JWT_EXPIRES_IN = JWT_EXPIRES_IN -app.state.AUTH_TRUSTED_EMAIL_HEADER = WEBUI_AUTH_TRUSTED_EMAIL_HEADER -app.state.AUTH_TRUSTED_NAME_HEADER = WEBUI_AUTH_TRUSTED_NAME_HEADER - - -app.state.config.SHOW_ADMIN_DETAILS = SHOW_ADMIN_DETAILS -app.state.config.ADMIN_EMAIL = ADMIN_EMAIL - - -app.state.config.DEFAULT_MODELS = DEFAULT_MODELS -app.state.config.DEFAULT_PROMPT_SUGGESTIONS = DEFAULT_PROMPT_SUGGESTIONS -app.state.config.DEFAULT_USER_ROLE = DEFAULT_USER_ROLE - - -app.state.config.USER_PERMISSIONS = USER_PERMISSIONS -app.state.config.WEBHOOK_URL = WEBHOOK_URL -app.state.config.BANNERS = WEBUI_BANNERS -app.state.config.MODEL_ORDER_LIST = MODEL_ORDER_LIST - -app.state.config.ENABLE_COMMUNITY_SHARING = ENABLE_COMMUNITY_SHARING -app.state.config.ENABLE_MESSAGE_RATING = ENABLE_MESSAGE_RATING - -app.state.config.ENABLE_EVALUATION_ARENA_MODELS = ENABLE_EVALUATION_ARENA_MODELS -app.state.config.EVALUATION_ARENA_MODELS = EVALUATION_ARENA_MODELS - -app.state.config.OAUTH_USERNAME_CLAIM = OAUTH_USERNAME_CLAIM -app.state.config.OAUTH_PICTURE_CLAIM = OAUTH_PICTURE_CLAIM -app.state.config.OAUTH_EMAIL_CLAIM = OAUTH_EMAIL_CLAIM - -app.state.config.ENABLE_OAUTH_ROLE_MANAGEMENT = ENABLE_OAUTH_ROLE_MANAGEMENT -app.state.config.OAUTH_ROLES_CLAIM = OAUTH_ROLES_CLAIM -app.state.config.OAUTH_ALLOWED_ROLES = OAUTH_ALLOWED_ROLES -app.state.config.OAUTH_ADMIN_ROLES = OAUTH_ADMIN_ROLES - -app.state.config.ENABLE_LDAP = ENABLE_LDAP -app.state.config.LDAP_SERVER_LABEL = LDAP_SERVER_LABEL -app.state.config.LDAP_SERVER_HOST = LDAP_SERVER_HOST -app.state.config.LDAP_SERVER_PORT = LDAP_SERVER_PORT -app.state.config.LDAP_ATTRIBUTE_FOR_USERNAME = LDAP_ATTRIBUTE_FOR_USERNAME -app.state.config.LDAP_APP_DN = LDAP_APP_DN -app.state.config.LDAP_APP_PASSWORD = LDAP_APP_PASSWORD -app.state.config.LDAP_SEARCH_BASE = LDAP_SEARCH_BASE -app.state.config.LDAP_SEARCH_FILTERS = LDAP_SEARCH_FILTERS -app.state.config.LDAP_USE_TLS = LDAP_USE_TLS -app.state.config.LDAP_CA_CERT_FILE = LDAP_CA_CERT_FILE -app.state.config.LDAP_CIPHERS = LDAP_CIPHERS - -app.state.TOOLS = {} -app.state.FUNCTIONS = {} - -app.add_middleware( - CORSMiddleware, - allow_origins=CORS_ALLOW_ORIGIN, - allow_credentials=True, - allow_methods=["*"], - allow_headers=["*"], -) - - -app.include_router(configs.router, prefix="/configs", tags=["configs"]) - -app.include_router(auths.router, prefix="/auths", tags=["auths"]) -app.include_router(users.router, prefix="/users", tags=["users"]) - -app.include_router(chats.router, prefix="/chats", tags=["chats"]) - -app.include_router(models.router, prefix="/models", tags=["models"]) -app.include_router(knowledge.router, prefix="/knowledge", tags=["knowledge"]) -app.include_router(prompts.router, prefix="/prompts", tags=["prompts"]) -app.include_router(tools.router, prefix="/tools", tags=["tools"]) - -app.include_router(memories.router, prefix="/memories", tags=["memories"]) -app.include_router(folders.router, prefix="/folders", tags=["folders"]) - -app.include_router(groups.router, prefix="/groups", tags=["groups"]) -app.include_router(files.router, prefix="/files", tags=["files"]) -app.include_router(functions.router, prefix="/functions", tags=["functions"]) -app.include_router(evaluations.router, prefix="/evaluations", tags=["evaluations"]) - - -app.include_router(utils.router, prefix="/utils", tags=["utils"]) - - -@app.get("/") -async def get_status(): - return { - "status": True, - "auth": WEBUI_AUTH, - "default_models": app.state.config.DEFAULT_MODELS, - "default_prompt_suggestions": app.state.config.DEFAULT_PROMPT_SUGGESTIONS, - } - - -async def get_all_models(): - models = [] - pipe_models = await get_pipe_models() - models = models + pipe_models - - if app.state.config.ENABLE_EVALUATION_ARENA_MODELS: - arena_models = [] - if len(app.state.config.EVALUATION_ARENA_MODELS) > 0: - arena_models = [ - { - "id": model["id"], - "name": model["name"], - "info": { - "meta": model["meta"], - }, - "object": "model", - "created": int(time.time()), - "owned_by": "arena", - "arena": True, - } - for model in app.state.config.EVALUATION_ARENA_MODELS - ] - else: - # Add default arena model - arena_models = [ - { - "id": DEFAULT_ARENA_MODEL["id"], - "name": DEFAULT_ARENA_MODEL["name"], - "info": { - "meta": DEFAULT_ARENA_MODEL["meta"], - }, - "object": "model", - "created": int(time.time()), - "owned_by": "arena", - "arena": True, - } - ] - models = models + arena_models - return models - - -def get_function_module(pipe_id: str): - # Check if function is already loaded - if pipe_id not in app.state.FUNCTIONS: - function_module, _, _ = load_function_module_by_id(pipe_id) - app.state.FUNCTIONS[pipe_id] = function_module - else: - function_module = app.state.FUNCTIONS[pipe_id] - - if hasattr(function_module, "valves") and hasattr(function_module, "Valves"): - valves = Functions.get_function_valves_by_id(pipe_id) - function_module.valves = function_module.Valves(**(valves if valves else {})) - return function_module - - -async def get_pipe_models(): - pipes = Functions.get_functions_by_type("pipe", active_only=True) - pipe_models = [] - - for pipe in pipes: - function_module = get_function_module(pipe.id) - - # Check if function is a manifold - if hasattr(function_module, "pipes"): - sub_pipes = [] - - # Check if pipes is a function or a list - - try: - if callable(function_module.pipes): - sub_pipes = function_module.pipes() - else: - sub_pipes = function_module.pipes - except Exception as e: - log.exception(e) - sub_pipes = [] - - log.debug( - f"get_pipe_models: function '{pipe.id}' is a manifold of {sub_pipes}" - ) - - for p in sub_pipes: - sub_pipe_id = f'{pipe.id}.{p["id"]}' - sub_pipe_name = p["name"] - - if hasattr(function_module, "name"): - sub_pipe_name = f"{function_module.name}{sub_pipe_name}" - - pipe_flag = {"type": pipe.type} - - pipe_models.append( - { - "id": sub_pipe_id, - "name": sub_pipe_name, - "object": "model", - "created": pipe.created_at, - "owned_by": "openai", - "pipe": pipe_flag, - } - ) - else: - pipe_flag = {"type": "pipe"} - - log.debug( - f"get_pipe_models: function '{pipe.id}' is a single pipe {{ 'id': {pipe.id}, 'name': {pipe.name} }}" - ) - - pipe_models.append( - { - "id": pipe.id, - "name": pipe.name, - "object": "model", - "created": pipe.created_at, - "owned_by": "openai", - "pipe": pipe_flag, - } - ) - - return pipe_models - - -async def execute_pipe(pipe, params): - if inspect.iscoroutinefunction(pipe): - return await pipe(**params) - else: - return pipe(**params) - - -async def get_message_content(res: str | Generator | AsyncGenerator) -> str: - if isinstance(res, str): - return res - if isinstance(res, Generator): - return "".join(map(str, res)) - if isinstance(res, AsyncGenerator): - return "".join([str(stream) async for stream in res]) - - -def process_line(form_data: dict, line): - if isinstance(line, BaseModel): - line = line.model_dump_json() - line = f"data: {line}" - if isinstance(line, dict): - line = f"data: {json.dumps(line)}" - - try: - line = line.decode("utf-8") - except Exception: - pass - - if line.startswith("data:"): - return f"{line}\n\n" - else: - line = openai_chat_chunk_message_template(form_data["model"], line) - return f"data: {json.dumps(line)}\n\n" - - -def get_pipe_id(form_data: dict) -> str: - pipe_id = form_data["model"] - if "." in pipe_id: - pipe_id, _ = pipe_id.split(".", 1) - - return pipe_id - - -def get_function_params(function_module, form_data, user, extra_params=None): - if extra_params is None: - extra_params = {} - - pipe_id = get_pipe_id(form_data) - - # Get the signature of the function - sig = inspect.signature(function_module.pipe) - params = {"body": form_data} | { - k: v for k, v in extra_params.items() if k in sig.parameters - } - - if "__user__" in params and hasattr(function_module, "UserValves"): - user_valves = Functions.get_user_valves_by_id_and_user_id(pipe_id, user.id) - try: - params["__user__"]["valves"] = function_module.UserValves(**user_valves) - except Exception as e: - log.exception(e) - params["__user__"]["valves"] = function_module.UserValves() - - return params - - -async def generate_function_chat_completion(form_data, user, models: dict = {}): - model_id = form_data.get("model") - model_info = Models.get_model_by_id(model_id) - - metadata = form_data.pop("metadata", {}) - - files = metadata.get("files", []) - tool_ids = metadata.get("tool_ids", []) - # Check if tool_ids is None - if tool_ids is None: - tool_ids = [] - - __event_emitter__ = None - __event_call__ = None - __task__ = None - __task_body__ = None - - if metadata: - if all(k in metadata for k in ("session_id", "chat_id", "message_id")): - __event_emitter__ = get_event_emitter(metadata) - __event_call__ = get_event_call(metadata) - __task__ = metadata.get("task", None) - __task_body__ = metadata.get("task_body", None) - - extra_params = { - "__event_emitter__": __event_emitter__, - "__event_call__": __event_call__, - "__task__": __task__, - "__task_body__": __task_body__, - "__files__": files, - "__user__": { - "id": user.id, - "email": user.email, - "name": user.name, - "role": user.role, - }, - "__metadata__": metadata, - } - extra_params["__tools__"] = get_tools( - app, - tool_ids, - user, - { - **extra_params, - "__model__": models.get(form_data["model"], None), - "__messages__": form_data["messages"], - "__files__": files, - }, - ) - - if model_info: - if model_info.base_model_id: - form_data["model"] = model_info.base_model_id - - params = model_info.params.model_dump() - form_data = apply_model_params_to_body_openai(params, form_data) - form_data = apply_model_system_prompt_to_body(params, form_data, user) - - pipe_id = get_pipe_id(form_data) - function_module = get_function_module(pipe_id) - - pipe = function_module.pipe - params = get_function_params(function_module, form_data, user, extra_params) - - if form_data.get("stream", False): - - async def stream_content(): - try: - res = await execute_pipe(pipe, params) - - # Directly return if the response is a StreamingResponse - if isinstance(res, StreamingResponse): - async for data in res.body_iterator: - yield data - return - if isinstance(res, dict): - yield f"data: {json.dumps(res)}\n\n" - return - - except Exception as e: - log.error(f"Error: {e}") - yield f"data: {json.dumps({'error': {'detail':str(e)}})}\n\n" - return - - if isinstance(res, str): - message = openai_chat_chunk_message_template(form_data["model"], res) - yield f"data: {json.dumps(message)}\n\n" - - if isinstance(res, Iterator): - for line in res: - yield process_line(form_data, line) - - if isinstance(res, AsyncGenerator): - async for line in res: - yield process_line(form_data, line) - - if isinstance(res, str) or isinstance(res, Generator): - finish_message = openai_chat_chunk_message_template( - form_data["model"], "" - ) - finish_message["choices"][0]["finish_reason"] = "stop" - yield f"data: {json.dumps(finish_message)}\n\n" - yield "data: [DONE]" - - return StreamingResponse(stream_content(), media_type="text/event-stream") - else: - try: - res = await execute_pipe(pipe, params) - - except Exception as e: - log.error(f"Error: {e}") - return {"error": {"detail": str(e)}} - - if isinstance(res, StreamingResponse) or isinstance(res, dict): - return res - if isinstance(res, BaseModel): - return res.model_dump() - - message = await get_message_content(res) - return openai_chat_completion_message_template(form_data["model"], message) diff --git a/backend/open_webui/config.py b/backend/open_webui/config.py index 905d2472a..e49c251a1 100644 --- a/backend/open_webui/config.py +++ b/backend/open_webui/config.py @@ -10,7 +10,7 @@ from urllib.parse import urlparse import chromadb import requests import yaml -from open_webui.apps.webui.internal.db import Base, get_db +from open_webui.internal.db import Base, get_db from open_webui.env import ( OPEN_WEBUI_DIR, DATA_DIR, @@ -21,6 +21,7 @@ from open_webui.env import ( WEBUI_NAME, log, DATABASE_URL, + OFFLINE_MODE ) from pydantic import BaseModel from sqlalchemy import JSON, Column, DateTime, Integer, func @@ -432,7 +433,10 @@ OAUTH_ADMIN_ROLES = PersistentConfig( OAUTH_ALLOWED_DOMAINS = PersistentConfig( "OAUTH_ALLOWED_DOMAINS", "oauth.allowed_domains", - [domain.strip() for domain in os.environ.get("OAUTH_ALLOWED_DOMAINS", "*").split(",")], + [ + domain.strip() + for domain in os.environ.get("OAUTH_ALLOWED_DOMAINS", "*").split(",") + ], ) @@ -954,12 +958,45 @@ TITLE_GENERATION_PROMPT_TEMPLATE = PersistentConfig( os.environ.get("TITLE_GENERATION_PROMPT_TEMPLATE", ""), ) +DEFAULT_TITLE_GENERATION_PROMPT_TEMPLATE = """Create a concise, 3-5 word title with an emoji as a title for the chat history, in the given language. Suitable Emojis for the summary can be used to enhance understanding but avoid quotation marks or special formatting. RESPOND ONLY WITH THE TITLE TEXT. + +Examples of titles: +📉 Stock Market Trends +🍪 Perfect Chocolate Chip Recipe +Evolution of Music Streaming +Remote Work Productivity Tips +Artificial Intelligence in Healthcare +🎮 Video Game Development Insights + + +{{MESSAGES:END:2}} +""" + + TAGS_GENERATION_PROMPT_TEMPLATE = PersistentConfig( "TAGS_GENERATION_PROMPT_TEMPLATE", "task.tags.prompt_template", os.environ.get("TAGS_GENERATION_PROMPT_TEMPLATE", ""), ) +DEFAULT_TAGS_GENERATION_PROMPT_TEMPLATE = """### Task: +Generate 1-3 broad tags categorizing the main themes of the chat history, along with 1-3 more specific subtopic tags. + +### Guidelines: +- Start with high-level domains (e.g. Science, Technology, Philosophy, Arts, Politics, Business, Health, Sports, Entertainment, Education) +- Consider including relevant subfields/subdomains if they are strongly represented throughout the conversation +- If content is too short (less than 3 messages) or too diverse, use only ["General"] +- Use the chat's primary language; default to English if multilingual +- Prioritize accuracy over specificity + +### Output: +JSON format: { "tags": ["tag1", "tag2", "tag3"] } + +### Chat History: + +{{MESSAGES:END:6}} +""" + ENABLE_TAGS_GENERATION = PersistentConfig( "ENABLE_TAGS_GENERATION", "task.tags.enable", @@ -1078,6 +1115,19 @@ TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE = PersistentConfig( ) +DEFAULT_TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE = """Available Tools: {{TOOLS}}\nReturn an empty string if no tools match the query. If a function tool matches, construct and return a JSON object in the format {\"name\": \"functionName\", \"parameters\": {\"requiredFunctionParamKey\": \"requiredFunctionParamValue\"}} using the appropriate tool and its parameters. Only return the object and limit the response to the JSON object without additional text.""" + + +DEFAULT_EMOJI_GENERATION_PROMPT_TEMPLATE = """Your task is to reflect the speaker's likely facial expression through a fitting emoji. Interpret emotions from the message and reflect their facial expression using fitting, diverse emojis (e.g., 😊, 😢, 😡, 😱). + +Message: ```{{prompt}}```""" + +DEFAULT_MOA_GENERATION_PROMPT_TEMPLATE = """You have been provided with a set of responses from various models to the latest user query: "{{prompt}}" + +Your task is to synthesize these responses into a single, high-quality response. It is crucial to critically evaluate the information provided in these responses, recognizing that some of it may be biased or incorrect. Your response should not simply replicate the given answers but should offer a refined, accurate, and comprehensive reply to the instruction. Ensure your response is well-structured, coherent, and adheres to the highest standards of accuracy and reliability. + +Responses from models: {{responses}}""" + #################################### # Vector Database #################################### @@ -1203,7 +1253,7 @@ RAG_EMBEDDING_MODEL = PersistentConfig( log.info(f"Embedding model set: {RAG_EMBEDDING_MODEL.value}") RAG_EMBEDDING_MODEL_AUTO_UPDATE = ( - os.environ.get("RAG_EMBEDDING_MODEL_AUTO_UPDATE", "True").lower() == "true" + not OFFLINE_MODE and os.environ.get("RAG_EMBEDDING_MODEL_AUTO_UPDATE", "True").lower() == "true" ) RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE = ( @@ -1228,7 +1278,7 @@ if RAG_RERANKING_MODEL.value != "": log.info(f"Reranking model set: {RAG_RERANKING_MODEL.value}") RAG_RERANKING_MODEL_AUTO_UPDATE = ( - os.environ.get("RAG_RERANKING_MODEL_AUTO_UPDATE", "True").lower() == "true" + not OFFLINE_MODE and os.environ.get("RAG_RERANKING_MODEL_AUTO_UPDATE", "True").lower() == "true" ) RAG_RERANKING_MODEL_TRUST_REMOTE_CODE = ( @@ -1698,7 +1748,7 @@ WHISPER_MODEL = PersistentConfig( WHISPER_MODEL_DIR = os.getenv("WHISPER_MODEL_DIR", f"{CACHE_DIR}/whisper/models") WHISPER_MODEL_AUTO_UPDATE = ( - os.environ.get("WHISPER_MODEL_AUTO_UPDATE", "").lower() == "true" + not OFFLINE_MODE and os.environ.get("WHISPER_MODEL_AUTO_UPDATE", "").lower() == "true" ) diff --git a/backend/open_webui/env.py b/backend/open_webui/env.py index e1b350ead..0fd6080de 100644 --- a/backend/open_webui/env.py +++ b/backend/open_webui/env.py @@ -376,7 +376,7 @@ else: AIOHTTP_CLIENT_TIMEOUT = 300 AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST = os.environ.get( - "AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST", "5" + "AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST", "" ) if AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST == "": diff --git a/backend/open_webui/functions.py b/backend/open_webui/functions.py new file mode 100644 index 000000000..9c241432a --- /dev/null +++ b/backend/open_webui/functions.py @@ -0,0 +1,316 @@ +import logging +import sys +import inspect +import json + +from pydantic import BaseModel +from typing import AsyncGenerator, Generator, Iterator +from fastapi import ( + Depends, + FastAPI, + File, + Form, + HTTPException, + Request, + UploadFile, + status, +) +from starlette.responses import Response, StreamingResponse + + +from open_webui.socket.main import ( + get_event_call, + get_event_emitter, +) + + +from open_webui.models.functions import Functions +from open_webui.models.models import Models + +from open_webui.utils.plugin import load_function_module_by_id +from open_webui.utils.tools import get_tools +from open_webui.utils.access_control import has_access + +from open_webui.env import SRC_LOG_LEVELS, GLOBAL_LOG_LEVEL + +from open_webui.utils.misc import ( + add_or_update_system_message, + get_last_user_message, + prepend_to_first_user_message_content, + openai_chat_chunk_message_template, + openai_chat_completion_message_template, +) +from open_webui.utils.payload import ( + apply_model_params_to_body_openai, + apply_model_system_prompt_to_body, +) + + +logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL) +log = logging.getLogger(__name__) +log.setLevel(SRC_LOG_LEVELS["MAIN"]) + + +def get_function_module_by_id(request: Request, pipe_id: str): + # Check if function is already loaded + if pipe_id not in request.app.state.FUNCTIONS: + function_module, _, _ = load_function_module_by_id(pipe_id) + request.app.state.FUNCTIONS[pipe_id] = function_module + else: + function_module = request.app.state.FUNCTIONS[pipe_id] + + if hasattr(function_module, "valves") and hasattr(function_module, "Valves"): + valves = Functions.get_function_valves_by_id(pipe_id) + function_module.valves = function_module.Valves(**(valves if valves else {})) + return function_module + + +async def get_function_models(): + pipes = Functions.get_functions_by_type("pipe", active_only=True) + pipe_models = [] + + for pipe in pipes: + function_module = get_function_module_by_id(pipe.id) + + # Check if function is a manifold + if hasattr(function_module, "pipes"): + sub_pipes = [] + + # Check if pipes is a function or a list + + try: + if callable(function_module.pipes): + sub_pipes = function_module.pipes() + else: + sub_pipes = function_module.pipes + except Exception as e: + log.exception(e) + sub_pipes = [] + + log.debug( + f"get_function_models: function '{pipe.id}' is a manifold of {sub_pipes}" + ) + + for p in sub_pipes: + sub_pipe_id = f'{pipe.id}.{p["id"]}' + sub_pipe_name = p["name"] + + if hasattr(function_module, "name"): + sub_pipe_name = f"{function_module.name}{sub_pipe_name}" + + pipe_flag = {"type": pipe.type} + + pipe_models.append( + { + "id": sub_pipe_id, + "name": sub_pipe_name, + "object": "model", + "created": pipe.created_at, + "owned_by": "openai", + "pipe": pipe_flag, + } + ) + else: + pipe_flag = {"type": "pipe"} + + log.debug( + f"get_function_models: function '{pipe.id}' is a single pipe {{ 'id': {pipe.id}, 'name': {pipe.name} }}" + ) + + pipe_models.append( + { + "id": pipe.id, + "name": pipe.name, + "object": "model", + "created": pipe.created_at, + "owned_by": "openai", + "pipe": pipe_flag, + } + ) + + return pipe_models + + +async def generate_function_chat_completion( + request, form_data, user, models: dict = {} +): + async def execute_pipe(pipe, params): + if inspect.iscoroutinefunction(pipe): + return await pipe(**params) + else: + return pipe(**params) + + async def get_message_content(res: str | Generator | AsyncGenerator) -> str: + if isinstance(res, str): + return res + if isinstance(res, Generator): + return "".join(map(str, res)) + if isinstance(res, AsyncGenerator): + return "".join([str(stream) async for stream in res]) + + def process_line(form_data: dict, line): + if isinstance(line, BaseModel): + line = line.model_dump_json() + line = f"data: {line}" + if isinstance(line, dict): + line = f"data: {json.dumps(line)}" + + try: + line = line.decode("utf-8") + except Exception: + pass + + if line.startswith("data:"): + return f"{line}\n\n" + else: + line = openai_chat_chunk_message_template(form_data["model"], line) + return f"data: {json.dumps(line)}\n\n" + + def get_pipe_id(form_data: dict) -> str: + pipe_id = form_data["model"] + if "." in pipe_id: + pipe_id, _ = pipe_id.split(".", 1) + return pipe_id + + def get_function_params(function_module, form_data, user, extra_params=None): + if extra_params is None: + extra_params = {} + + pipe_id = get_pipe_id(form_data) + + # Get the signature of the function + sig = inspect.signature(function_module.pipe) + params = {"body": form_data} | { + k: v for k, v in extra_params.items() if k in sig.parameters + } + + if "__user__" in params and hasattr(function_module, "UserValves"): + user_valves = Functions.get_user_valves_by_id_and_user_id(pipe_id, user.id) + try: + params["__user__"]["valves"] = function_module.UserValves(**user_valves) + except Exception as e: + log.exception(e) + params["__user__"]["valves"] = function_module.UserValves() + + return params + + model_id = form_data.get("model") + model_info = Models.get_model_by_id(model_id) + + metadata = form_data.pop("metadata", {}) + + files = metadata.get("files", []) + tool_ids = metadata.get("tool_ids", []) + # Check if tool_ids is None + if tool_ids is None: + tool_ids = [] + + __event_emitter__ = None + __event_call__ = None + __task__ = None + __task_body__ = None + + if metadata: + if all(k in metadata for k in ("session_id", "chat_id", "message_id")): + __event_emitter__ = get_event_emitter(metadata) + __event_call__ = get_event_call(metadata) + __task__ = metadata.get("task", None) + __task_body__ = metadata.get("task_body", None) + + extra_params = { + "__event_emitter__": __event_emitter__, + "__event_call__": __event_call__, + "__task__": __task__, + "__task_body__": __task_body__, + "__files__": files, + "__user__": { + "id": user.id, + "email": user.email, + "name": user.name, + "role": user.role, + }, + "__metadata__": metadata, + "__request__": request, + } + extra_params["__tools__"] = get_tools( + request, + tool_ids, + user, + { + **extra_params, + "__model__": models.get(form_data["model"], None), + "__messages__": form_data["messages"], + "__files__": files, + }, + ) + + if model_info: + if model_info.base_model_id: + form_data["model"] = model_info.base_model_id + + params = model_info.params.model_dump() + form_data = apply_model_params_to_body_openai(params, form_data) + form_data = apply_model_system_prompt_to_body(params, form_data, user) + + pipe_id = get_pipe_id(form_data) + function_module = get_function_module_by_id(pipe_id) + + pipe = function_module.pipe + params = get_function_params(function_module, form_data, user, extra_params) + + if form_data.get("stream", False): + + async def stream_content(): + try: + res = await execute_pipe(pipe, params) + + # Directly return if the response is a StreamingResponse + if isinstance(res, StreamingResponse): + async for data in res.body_iterator: + yield data + return + if isinstance(res, dict): + yield f"data: {json.dumps(res)}\n\n" + return + + except Exception as e: + log.error(f"Error: {e}") + yield f"data: {json.dumps({'error': {'detail':str(e)}})}\n\n" + return + + if isinstance(res, str): + message = openai_chat_chunk_message_template(form_data["model"], res) + yield f"data: {json.dumps(message)}\n\n" + + if isinstance(res, Iterator): + for line in res: + yield process_line(form_data, line) + + if isinstance(res, AsyncGenerator): + async for line in res: + yield process_line(form_data, line) + + if isinstance(res, str) or isinstance(res, Generator): + finish_message = openai_chat_chunk_message_template( + form_data["model"], "" + ) + finish_message["choices"][0]["finish_reason"] = "stop" + yield f"data: {json.dumps(finish_message)}\n\n" + yield "data: [DONE]" + + return StreamingResponse(stream_content(), media_type="text/event-stream") + else: + try: + res = await execute_pipe(pipe, params) + + except Exception as e: + log.error(f"Error: {e}") + return {"error": {"detail": str(e)}} + + if isinstance(res, StreamingResponse) or isinstance(res, dict): + return res + if isinstance(res, BaseModel): + return res.model_dump() + + message = await get_message_content(res) + return openai_chat_completion_message_template(form_data["model"], message) diff --git a/backend/open_webui/apps/webui/internal/db.py b/backend/open_webui/internal/db.py similarity index 97% rename from backend/open_webui/apps/webui/internal/db.py rename to backend/open_webui/internal/db.py index bcf913e6f..ba078822e 100644 --- a/backend/open_webui/apps/webui/internal/db.py +++ b/backend/open_webui/internal/db.py @@ -3,7 +3,7 @@ import logging from contextlib import contextmanager from typing import Any, Optional -from open_webui.apps.webui.internal.wrappers import register_connection +from open_webui.internal.wrappers import register_connection from open_webui.env import ( OPEN_WEBUI_DIR, DATABASE_URL, diff --git a/backend/open_webui/apps/webui/internal/migrations/001_initial_schema.py b/backend/open_webui/internal/migrations/001_initial_schema.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/001_initial_schema.py rename to backend/open_webui/internal/migrations/001_initial_schema.py diff --git a/backend/open_webui/apps/webui/internal/migrations/002_add_local_sharing.py b/backend/open_webui/internal/migrations/002_add_local_sharing.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/002_add_local_sharing.py rename to backend/open_webui/internal/migrations/002_add_local_sharing.py diff --git a/backend/open_webui/apps/webui/internal/migrations/003_add_auth_api_key.py b/backend/open_webui/internal/migrations/003_add_auth_api_key.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/003_add_auth_api_key.py rename to backend/open_webui/internal/migrations/003_add_auth_api_key.py diff --git a/backend/open_webui/apps/webui/internal/migrations/004_add_archived.py b/backend/open_webui/internal/migrations/004_add_archived.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/004_add_archived.py rename to backend/open_webui/internal/migrations/004_add_archived.py diff --git a/backend/open_webui/apps/webui/internal/migrations/005_add_updated_at.py b/backend/open_webui/internal/migrations/005_add_updated_at.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/005_add_updated_at.py rename to backend/open_webui/internal/migrations/005_add_updated_at.py diff --git a/backend/open_webui/apps/webui/internal/migrations/006_migrate_timestamps_and_charfields.py b/backend/open_webui/internal/migrations/006_migrate_timestamps_and_charfields.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/006_migrate_timestamps_and_charfields.py rename to backend/open_webui/internal/migrations/006_migrate_timestamps_and_charfields.py diff --git a/backend/open_webui/apps/webui/internal/migrations/007_add_user_last_active_at.py b/backend/open_webui/internal/migrations/007_add_user_last_active_at.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/007_add_user_last_active_at.py rename to backend/open_webui/internal/migrations/007_add_user_last_active_at.py diff --git a/backend/open_webui/apps/webui/internal/migrations/008_add_memory.py b/backend/open_webui/internal/migrations/008_add_memory.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/008_add_memory.py rename to backend/open_webui/internal/migrations/008_add_memory.py diff --git a/backend/open_webui/apps/webui/internal/migrations/009_add_models.py b/backend/open_webui/internal/migrations/009_add_models.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/009_add_models.py rename to backend/open_webui/internal/migrations/009_add_models.py diff --git a/backend/open_webui/apps/webui/internal/migrations/010_migrate_modelfiles_to_models.py b/backend/open_webui/internal/migrations/010_migrate_modelfiles_to_models.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/010_migrate_modelfiles_to_models.py rename to backend/open_webui/internal/migrations/010_migrate_modelfiles_to_models.py diff --git a/backend/open_webui/apps/webui/internal/migrations/011_add_user_settings.py b/backend/open_webui/internal/migrations/011_add_user_settings.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/011_add_user_settings.py rename to backend/open_webui/internal/migrations/011_add_user_settings.py diff --git a/backend/open_webui/apps/webui/internal/migrations/012_add_tools.py b/backend/open_webui/internal/migrations/012_add_tools.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/012_add_tools.py rename to backend/open_webui/internal/migrations/012_add_tools.py diff --git a/backend/open_webui/apps/webui/internal/migrations/013_add_user_info.py b/backend/open_webui/internal/migrations/013_add_user_info.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/013_add_user_info.py rename to backend/open_webui/internal/migrations/013_add_user_info.py diff --git a/backend/open_webui/apps/webui/internal/migrations/014_add_files.py b/backend/open_webui/internal/migrations/014_add_files.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/014_add_files.py rename to backend/open_webui/internal/migrations/014_add_files.py diff --git a/backend/open_webui/apps/webui/internal/migrations/015_add_functions.py b/backend/open_webui/internal/migrations/015_add_functions.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/015_add_functions.py rename to backend/open_webui/internal/migrations/015_add_functions.py diff --git a/backend/open_webui/apps/webui/internal/migrations/016_add_valves_and_is_active.py b/backend/open_webui/internal/migrations/016_add_valves_and_is_active.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/016_add_valves_and_is_active.py rename to backend/open_webui/internal/migrations/016_add_valves_and_is_active.py diff --git a/backend/open_webui/apps/webui/internal/migrations/017_add_user_oauth_sub.py b/backend/open_webui/internal/migrations/017_add_user_oauth_sub.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/017_add_user_oauth_sub.py rename to backend/open_webui/internal/migrations/017_add_user_oauth_sub.py diff --git a/backend/open_webui/apps/webui/internal/migrations/018_add_function_is_global.py b/backend/open_webui/internal/migrations/018_add_function_is_global.py similarity index 100% rename from backend/open_webui/apps/webui/internal/migrations/018_add_function_is_global.py rename to backend/open_webui/internal/migrations/018_add_function_is_global.py diff --git a/backend/open_webui/apps/webui/internal/wrappers.py b/backend/open_webui/internal/wrappers.py similarity index 100% rename from backend/open_webui/apps/webui/internal/wrappers.py rename to backend/open_webui/internal/wrappers.py diff --git a/backend/open_webui/main.py b/backend/open_webui/main.py index 253a7a165..31604984f 100644 --- a/backend/open_webui/main.py +++ b/backend/open_webui/main.py @@ -8,9 +8,13 @@ import shutil import sys import time import random -from contextlib import asynccontextmanager -from typing import Optional +from contextlib import asynccontextmanager +from urllib.parse import urlencode, parse_qs, urlparse +from pydantic import BaseModel +from sqlalchemy import text + +from typing import Optional from aiocache import cached import aiohttp import requests @@ -27,126 +31,260 @@ from fastapi import ( from fastapi.middleware.cors import CORSMiddleware from fastapi.responses import JSONResponse, RedirectResponse from fastapi.staticfiles import StaticFiles -from pydantic import BaseModel -from sqlalchemy import text + from starlette.exceptions import HTTPException as StarletteHTTPException from starlette.middleware.base import BaseHTTPMiddleware from starlette.middleware.sessions import SessionMiddleware from starlette.responses import Response, StreamingResponse -from open_webui.apps.audio.main import app as audio_app -from open_webui.apps.images.main import app as images_app -from open_webui.apps.ollama.main import ( - app as ollama_app, - get_all_models as get_ollama_models, - generate_chat_completion as generate_ollama_chat_completion, - GenerateChatCompletionForm, -) -from open_webui.apps.openai.main import ( - app as openai_app, - generate_chat_completion as generate_openai_chat_completion, - get_all_models as get_openai_models, - get_all_models_responses as get_openai_models_responses, -) -from open_webui.apps.retrieval.main import app as retrieval_app -from open_webui.apps.retrieval.utils import get_sources_from_files - -from open_webui.apps.socket.main import ( +from open_webui.socket.main import ( app as socket_app, periodic_usage_pool_cleanup, - get_event_call, - get_event_emitter, ) -from open_webui.apps.webui.internal.db import Session -from open_webui.apps.webui.main import ( - app as webui_app, - generate_function_chat_completion, - get_all_models as get_open_webui_models, +from open_webui.routers import ( + audio, + images, + ollama, + openai, + retrieval, + pipelines, + tasks, + auths, + chats, + folders, + configs, + groups, + files, + functions, + memories, + models, + knowledge, + prompts, + evaluations, + tools, + users, + utils, ) -from open_webui.apps.webui.models.functions import Functions -from open_webui.apps.webui.models.models import Models -from open_webui.apps.webui.models.users import UserModel, Users -from open_webui.apps.webui.utils import load_function_module_by_id + +from open_webui.routers.retrieval import ( + get_embedding_function, + get_ef, + get_rf, +) + +from open_webui.internal.db import Session + +from open_webui.models.functions import Functions +from open_webui.models.models import Models +from open_webui.models.users import UserModel, Users + from open_webui.config import ( - CACHE_DIR, - CORS_ALLOW_ORIGIN, - DEFAULT_LOCALE, - ENABLE_ADMIN_CHAT_ACCESS, - ENABLE_ADMIN_EXPORT, + # Ollama ENABLE_OLLAMA_API, + OLLAMA_BASE_URLS, + OLLAMA_API_CONFIGS, + # OpenAI ENABLE_OPENAI_API, - ENABLE_TAGS_GENERATION, - ENV, - FRONTEND_BUILD_DIR, - OAUTH_PROVIDERS, - STATIC_DIR, - TASK_MODEL, - TASK_MODEL_EXTERNAL, - ENABLE_SEARCH_QUERY_GENERATION, - ENABLE_RETRIEVAL_QUERY_GENERATION, - QUERY_GENERATION_PROMPT_TEMPLATE, - DEFAULT_QUERY_GENERATION_PROMPT_TEMPLATE, - TITLE_GENERATION_PROMPT_TEMPLATE, - TAGS_GENERATION_PROMPT_TEMPLATE, - ENABLE_AUTOCOMPLETE_GENERATION, - AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH, - AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE, - DEFAULT_AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE, - TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE, - WEBHOOK_URL, + OPENAI_API_BASE_URLS, + OPENAI_API_KEYS, + OPENAI_API_CONFIGS, + # Image + AUTOMATIC1111_API_AUTH, + AUTOMATIC1111_BASE_URL, + AUTOMATIC1111_CFG_SCALE, + AUTOMATIC1111_SAMPLER, + AUTOMATIC1111_SCHEDULER, + COMFYUI_BASE_URL, + COMFYUI_WORKFLOW, + COMFYUI_WORKFLOW_NODES, + ENABLE_IMAGE_GENERATION, + IMAGE_GENERATION_ENGINE, + IMAGE_GENERATION_MODEL, + IMAGE_SIZE, + IMAGE_STEPS, + IMAGES_OPENAI_API_BASE_URL, + IMAGES_OPENAI_API_KEY, + # Audio + AUDIO_STT_ENGINE, + AUDIO_STT_MODEL, + AUDIO_STT_OPENAI_API_BASE_URL, + AUDIO_STT_OPENAI_API_KEY, + AUDIO_TTS_API_KEY, + AUDIO_TTS_ENGINE, + AUDIO_TTS_MODEL, + AUDIO_TTS_OPENAI_API_BASE_URL, + AUDIO_TTS_OPENAI_API_KEY, + AUDIO_TTS_SPLIT_ON, + AUDIO_TTS_VOICE, + AUDIO_TTS_AZURE_SPEECH_REGION, + AUDIO_TTS_AZURE_SPEECH_OUTPUT_FORMAT, + WHISPER_MODEL, + WHISPER_MODEL_AUTO_UPDATE, + WHISPER_MODEL_DIR, + # Retrieval + RAG_TEMPLATE, + DEFAULT_RAG_TEMPLATE, + RAG_EMBEDDING_MODEL, + RAG_EMBEDDING_MODEL_AUTO_UPDATE, + RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE, + RAG_RERANKING_MODEL, + RAG_RERANKING_MODEL_AUTO_UPDATE, + RAG_RERANKING_MODEL_TRUST_REMOTE_CODE, + RAG_EMBEDDING_ENGINE, + RAG_EMBEDDING_BATCH_SIZE, + RAG_RELEVANCE_THRESHOLD, + RAG_FILE_MAX_COUNT, + RAG_FILE_MAX_SIZE, + RAG_OPENAI_API_BASE_URL, + RAG_OPENAI_API_KEY, + RAG_OLLAMA_BASE_URL, + RAG_OLLAMA_API_KEY, + CHUNK_OVERLAP, + CHUNK_SIZE, + CONTENT_EXTRACTION_ENGINE, + TIKA_SERVER_URL, + RAG_TOP_K, + RAG_TEXT_SPLITTER, + TIKTOKEN_ENCODING_NAME, + PDF_EXTRACT_IMAGES, + YOUTUBE_LOADER_LANGUAGE, + YOUTUBE_LOADER_PROXY_URL, + # Retrieval (Web Search) + RAG_WEB_SEARCH_ENGINE, + RAG_WEB_SEARCH_RESULT_COUNT, + RAG_WEB_SEARCH_CONCURRENT_REQUESTS, + RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + JINA_API_KEY, + SEARCHAPI_API_KEY, + SEARCHAPI_ENGINE, + SEARXNG_QUERY_URL, + SERPER_API_KEY, + SERPLY_API_KEY, + SERPSTACK_API_KEY, + SERPSTACK_HTTPS, + TAVILY_API_KEY, + BING_SEARCH_V7_ENDPOINT, + BING_SEARCH_V7_SUBSCRIPTION_KEY, + BRAVE_SEARCH_API_KEY, + KAGI_SEARCH_API_KEY, + MOJEEK_SEARCH_API_KEY, + GOOGLE_PSE_API_KEY, + GOOGLE_PSE_ENGINE_ID, + ENABLE_RAG_HYBRID_SEARCH, + ENABLE_RAG_LOCAL_WEB_FETCH, + ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION, + ENABLE_RAG_WEB_SEARCH, + UPLOAD_DIR, + # WebUI WEBUI_AUTH, WEBUI_NAME, + WEBUI_BANNERS, + WEBHOOK_URL, + ADMIN_EMAIL, + SHOW_ADMIN_DETAILS, + JWT_EXPIRES_IN, + ENABLE_SIGNUP, + ENABLE_LOGIN_FORM, + ENABLE_API_KEY, + ENABLE_COMMUNITY_SHARING, + ENABLE_MESSAGE_RATING, + ENABLE_EVALUATION_ARENA_MODELS, + USER_PERMISSIONS, + DEFAULT_USER_ROLE, + DEFAULT_PROMPT_SUGGESTIONS, + DEFAULT_MODELS, + DEFAULT_ARENA_MODEL, + MODEL_ORDER_LIST, + EVALUATION_ARENA_MODELS, + # WebUI (OAuth) + ENABLE_OAUTH_ROLE_MANAGEMENT, + OAUTH_ROLES_CLAIM, + OAUTH_EMAIL_CLAIM, + OAUTH_PICTURE_CLAIM, + OAUTH_USERNAME_CLAIM, + OAUTH_ALLOWED_ROLES, + OAUTH_ADMIN_ROLES, + # WebUI (LDAP) + ENABLE_LDAP, + LDAP_SERVER_LABEL, + LDAP_SERVER_HOST, + LDAP_SERVER_PORT, + LDAP_ATTRIBUTE_FOR_USERNAME, + LDAP_SEARCH_FILTERS, + LDAP_SEARCH_BASE, + LDAP_APP_DN, + LDAP_APP_PASSWORD, + LDAP_USE_TLS, + LDAP_CA_CERT_FILE, + LDAP_CIPHERS, + # Misc + ENV, + CACHE_DIR, + STATIC_DIR, + FRONTEND_BUILD_DIR, + CORS_ALLOW_ORIGIN, + DEFAULT_LOCALE, + OAUTH_PROVIDERS, + # Admin + ENABLE_ADMIN_CHAT_ACCESS, + ENABLE_ADMIN_EXPORT, + # Tasks + TASK_MODEL, + TASK_MODEL_EXTERNAL, + ENABLE_TAGS_GENERATION, + ENABLE_SEARCH_QUERY_GENERATION, + ENABLE_RETRIEVAL_QUERY_GENERATION, + ENABLE_AUTOCOMPLETE_GENERATION, + TITLE_GENERATION_PROMPT_TEMPLATE, + TAGS_GENERATION_PROMPT_TEMPLATE, + TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE, + QUERY_GENERATION_PROMPT_TEMPLATE, + AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE, + AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH, AppConfig, reset_config, ) -from open_webui.constants import TASKS from open_webui.env import ( CHANGELOG, GLOBAL_LOG_LEVEL, SAFE_MODE, SRC_LOG_LEVELS, VERSION, + WEBUI_URL, WEBUI_BUILD_HASH, WEBUI_SECRET_KEY, WEBUI_SESSION_COOKIE_SAME_SITE, WEBUI_SESSION_COOKIE_SECURE, - WEBUI_URL, + WEBUI_AUTH_TRUSTED_EMAIL_HEADER, + WEBUI_AUTH_TRUSTED_NAME_HEADER, BYPASS_MODEL_ACCESS_CONTROL, RESET_CONFIG_ON_START, OFFLINE_MODE, ) -from open_webui.utils.misc import ( - add_or_update_system_message, - get_last_user_message, - prepend_to_first_user_message_content, + + +from open_webui.utils.models import ( + get_all_models, + get_all_base_models, + check_model_access, ) -from open_webui.utils.oauth import oauth_manager -from open_webui.utils.payload import convert_payload_openai_to_ollama -from open_webui.utils.response import ( - convert_response_ollama_to_openai, - convert_streaming_response_ollama_to_openai, +from open_webui.utils.chat import ( + generate_chat_completion as chat_completion_handler, + chat_completed as chat_completed_handler, + chat_action as chat_action_handler, ) -from open_webui.utils.security_headers import SecurityHeadersMiddleware -from open_webui.utils.task import ( - rag_template, - title_generation_template, - query_generation_template, - autocomplete_generation_template, - tags_generation_template, - emoji_generation_template, - moa_response_generation_template, - tools_function_calling_generation_template, -) -from open_webui.utils.tools import get_tools +from open_webui.utils.middleware import process_chat_payload, process_chat_response +from open_webui.utils.access_control import has_access + from open_webui.utils.auth import ( decode_token, get_admin_user, - get_current_user, - get_http_authorization_cred, get_verified_user, ) -from open_webui.utils.access_control import has_access +from open_webui.utils.oauth import oauth_manager +from open_webui.utils.security_headers import SecurityHeadersMiddleware + if SAFE_MODE: print("SAFE MODE ENABLED") @@ -203,755 +341,298 @@ app = FastAPI( app.state.config = AppConfig() -app.state.config.ENABLE_OPENAI_API = ENABLE_OPENAI_API -app.state.config.ENABLE_OLLAMA_API = ENABLE_OLLAMA_API +######################################## +# +# OLLAMA +# +######################################## + + +app.state.config.ENABLE_OLLAMA_API = ENABLE_OLLAMA_API +app.state.config.OLLAMA_BASE_URLS = OLLAMA_BASE_URLS +app.state.config.OLLAMA_API_CONFIGS = OLLAMA_API_CONFIGS + +app.state.OLLAMA_MODELS = {} + +######################################## +# +# OPENAI +# +######################################## + +app.state.config.ENABLE_OPENAI_API = ENABLE_OPENAI_API +app.state.config.OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS +app.state.config.OPENAI_API_KEYS = OPENAI_API_KEYS +app.state.config.OPENAI_API_CONFIGS = OPENAI_API_CONFIGS + +app.state.OPENAI_MODELS = {} + +######################################## +# +# WEBUI +# +######################################## + +app.state.config.ENABLE_SIGNUP = ENABLE_SIGNUP +app.state.config.ENABLE_LOGIN_FORM = ENABLE_LOGIN_FORM +app.state.config.ENABLE_API_KEY = ENABLE_API_KEY + +app.state.config.JWT_EXPIRES_IN = JWT_EXPIRES_IN + +app.state.config.SHOW_ADMIN_DETAILS = SHOW_ADMIN_DETAILS +app.state.config.ADMIN_EMAIL = ADMIN_EMAIL + + +app.state.config.DEFAULT_MODELS = DEFAULT_MODELS +app.state.config.DEFAULT_PROMPT_SUGGESTIONS = DEFAULT_PROMPT_SUGGESTIONS +app.state.config.DEFAULT_USER_ROLE = DEFAULT_USER_ROLE + +app.state.config.USER_PERMISSIONS = USER_PERMISSIONS app.state.config.WEBHOOK_URL = WEBHOOK_URL +app.state.config.BANNERS = WEBUI_BANNERS +app.state.config.MODEL_ORDER_LIST = MODEL_ORDER_LIST + +app.state.config.ENABLE_COMMUNITY_SHARING = ENABLE_COMMUNITY_SHARING +app.state.config.ENABLE_MESSAGE_RATING = ENABLE_MESSAGE_RATING + +app.state.config.ENABLE_EVALUATION_ARENA_MODELS = ENABLE_EVALUATION_ARENA_MODELS +app.state.config.EVALUATION_ARENA_MODELS = EVALUATION_ARENA_MODELS + +app.state.config.OAUTH_USERNAME_CLAIM = OAUTH_USERNAME_CLAIM +app.state.config.OAUTH_PICTURE_CLAIM = OAUTH_PICTURE_CLAIM +app.state.config.OAUTH_EMAIL_CLAIM = OAUTH_EMAIL_CLAIM + +app.state.config.ENABLE_OAUTH_ROLE_MANAGEMENT = ENABLE_OAUTH_ROLE_MANAGEMENT +app.state.config.OAUTH_ROLES_CLAIM = OAUTH_ROLES_CLAIM +app.state.config.OAUTH_ALLOWED_ROLES = OAUTH_ALLOWED_ROLES +app.state.config.OAUTH_ADMIN_ROLES = OAUTH_ADMIN_ROLES + +app.state.config.ENABLE_LDAP = ENABLE_LDAP +app.state.config.LDAP_SERVER_LABEL = LDAP_SERVER_LABEL +app.state.config.LDAP_SERVER_HOST = LDAP_SERVER_HOST +app.state.config.LDAP_SERVER_PORT = LDAP_SERVER_PORT +app.state.config.LDAP_ATTRIBUTE_FOR_USERNAME = LDAP_ATTRIBUTE_FOR_USERNAME +app.state.config.LDAP_APP_DN = LDAP_APP_DN +app.state.config.LDAP_APP_PASSWORD = LDAP_APP_PASSWORD +app.state.config.LDAP_SEARCH_BASE = LDAP_SEARCH_BASE +app.state.config.LDAP_SEARCH_FILTERS = LDAP_SEARCH_FILTERS +app.state.config.LDAP_USE_TLS = LDAP_USE_TLS +app.state.config.LDAP_CA_CERT_FILE = LDAP_CA_CERT_FILE +app.state.config.LDAP_CIPHERS = LDAP_CIPHERS + + +app.state.AUTH_TRUSTED_EMAIL_HEADER = WEBUI_AUTH_TRUSTED_EMAIL_HEADER +app.state.AUTH_TRUSTED_NAME_HEADER = WEBUI_AUTH_TRUSTED_NAME_HEADER + +app.state.TOOLS = {} +app.state.FUNCTIONS = {} + + +######################################## +# +# RETRIEVAL +# +######################################## + + +app.state.config.TOP_K = RAG_TOP_K +app.state.config.RELEVANCE_THRESHOLD = RAG_RELEVANCE_THRESHOLD +app.state.config.FILE_MAX_SIZE = RAG_FILE_MAX_SIZE +app.state.config.FILE_MAX_COUNT = RAG_FILE_MAX_COUNT + +app.state.config.ENABLE_RAG_HYBRID_SEARCH = ENABLE_RAG_HYBRID_SEARCH +app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION = ( + ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION +) + +app.state.config.CONTENT_EXTRACTION_ENGINE = CONTENT_EXTRACTION_ENGINE +app.state.config.TIKA_SERVER_URL = TIKA_SERVER_URL + +app.state.config.TEXT_SPLITTER = RAG_TEXT_SPLITTER +app.state.config.TIKTOKEN_ENCODING_NAME = TIKTOKEN_ENCODING_NAME + +app.state.config.CHUNK_SIZE = CHUNK_SIZE +app.state.config.CHUNK_OVERLAP = CHUNK_OVERLAP + +app.state.config.RAG_EMBEDDING_ENGINE = RAG_EMBEDDING_ENGINE +app.state.config.RAG_EMBEDDING_MODEL = RAG_EMBEDDING_MODEL +app.state.config.RAG_EMBEDDING_BATCH_SIZE = RAG_EMBEDDING_BATCH_SIZE +app.state.config.RAG_RERANKING_MODEL = RAG_RERANKING_MODEL +app.state.config.RAG_TEMPLATE = RAG_TEMPLATE + +app.state.config.RAG_OPENAI_API_BASE_URL = RAG_OPENAI_API_BASE_URL +app.state.config.RAG_OPENAI_API_KEY = RAG_OPENAI_API_KEY + +app.state.config.RAG_OLLAMA_BASE_URL = RAG_OLLAMA_BASE_URL +app.state.config.RAG_OLLAMA_API_KEY = RAG_OLLAMA_API_KEY + +app.state.config.PDF_EXTRACT_IMAGES = PDF_EXTRACT_IMAGES + +app.state.config.YOUTUBE_LOADER_LANGUAGE = YOUTUBE_LOADER_LANGUAGE +app.state.config.YOUTUBE_LOADER_PROXY_URL = YOUTUBE_LOADER_PROXY_URL + + +app.state.config.ENABLE_RAG_WEB_SEARCH = ENABLE_RAG_WEB_SEARCH +app.state.config.RAG_WEB_SEARCH_ENGINE = RAG_WEB_SEARCH_ENGINE +app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST = RAG_WEB_SEARCH_DOMAIN_FILTER_LIST + +app.state.config.SEARXNG_QUERY_URL = SEARXNG_QUERY_URL +app.state.config.GOOGLE_PSE_API_KEY = GOOGLE_PSE_API_KEY +app.state.config.GOOGLE_PSE_ENGINE_ID = GOOGLE_PSE_ENGINE_ID +app.state.config.BRAVE_SEARCH_API_KEY = BRAVE_SEARCH_API_KEY +app.state.config.KAGI_SEARCH_API_KEY = KAGI_SEARCH_API_KEY +app.state.config.MOJEEK_SEARCH_API_KEY = MOJEEK_SEARCH_API_KEY +app.state.config.SERPSTACK_API_KEY = SERPSTACK_API_KEY +app.state.config.SERPSTACK_HTTPS = SERPSTACK_HTTPS +app.state.config.SERPER_API_KEY = SERPER_API_KEY +app.state.config.SERPLY_API_KEY = SERPLY_API_KEY +app.state.config.TAVILY_API_KEY = TAVILY_API_KEY +app.state.config.SEARCHAPI_API_KEY = SEARCHAPI_API_KEY +app.state.config.SEARCHAPI_ENGINE = SEARCHAPI_ENGINE +app.state.config.JINA_API_KEY = JINA_API_KEY +app.state.config.BING_SEARCH_V7_ENDPOINT = BING_SEARCH_V7_ENDPOINT +app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY = BING_SEARCH_V7_SUBSCRIPTION_KEY + +app.state.config.RAG_WEB_SEARCH_RESULT_COUNT = RAG_WEB_SEARCH_RESULT_COUNT +app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS = RAG_WEB_SEARCH_CONCURRENT_REQUESTS + +app.state.EMBEDDING_FUNCTION = None +app.state.ef = None +app.state.rf = None + +app.state.YOUTUBE_LOADER_TRANSLATION = None + + +app.state.EMBEDDING_FUNCTION = get_embedding_function( + app.state.config.RAG_EMBEDDING_ENGINE, + app.state.config.RAG_EMBEDDING_MODEL, + app.state.ef, + ( + app.state.config.RAG_OPENAI_API_BASE_URL + if app.state.config.RAG_EMBEDDING_ENGINE == "openai" + else app.state.config.RAG_OLLAMA_BASE_URL + ), + ( + app.state.config.RAG_OPENAI_API_KEY + if app.state.config.RAG_EMBEDDING_ENGINE == "openai" + else app.state.config.RAG_OLLAMA_API_KEY + ), + app.state.config.RAG_EMBEDDING_BATCH_SIZE, +) + +try: + app.state.ef = get_ef( + app.state.config.RAG_EMBEDDING_ENGINE, + app.state.config.RAG_EMBEDDING_MODEL, + RAG_EMBEDDING_MODEL_AUTO_UPDATE, + ) + + app.state.rf = get_rf( + app.state.config.RAG_RERANKING_MODEL, + RAG_RERANKING_MODEL_AUTO_UPDATE, + ) +except Exception as e: + log.error(f"Error updating models: {e}") + pass + + +######################################## +# +# IMAGES +# +######################################## + +app.state.config.IMAGE_GENERATION_ENGINE = IMAGE_GENERATION_ENGINE +app.state.config.ENABLE_IMAGE_GENERATION = ENABLE_IMAGE_GENERATION + +app.state.config.IMAGES_OPENAI_API_BASE_URL = IMAGES_OPENAI_API_BASE_URL +app.state.config.IMAGES_OPENAI_API_KEY = IMAGES_OPENAI_API_KEY + +app.state.config.IMAGE_GENERATION_MODEL = IMAGE_GENERATION_MODEL + +app.state.config.AUTOMATIC1111_BASE_URL = AUTOMATIC1111_BASE_URL +app.state.config.AUTOMATIC1111_API_AUTH = AUTOMATIC1111_API_AUTH +app.state.config.AUTOMATIC1111_CFG_SCALE = AUTOMATIC1111_CFG_SCALE +app.state.config.AUTOMATIC1111_SAMPLER = AUTOMATIC1111_SAMPLER +app.state.config.AUTOMATIC1111_SCHEDULER = AUTOMATIC1111_SCHEDULER +app.state.config.COMFYUI_BASE_URL = COMFYUI_BASE_URL +app.state.config.COMFYUI_WORKFLOW = COMFYUI_WORKFLOW +app.state.config.COMFYUI_WORKFLOW_NODES = COMFYUI_WORKFLOW_NODES + +app.state.config.IMAGE_SIZE = IMAGE_SIZE +app.state.config.IMAGE_STEPS = IMAGE_STEPS + + +######################################## +# +# AUDIO +# +######################################## + +app.state.config.STT_OPENAI_API_BASE_URL = AUDIO_STT_OPENAI_API_BASE_URL +app.state.config.STT_OPENAI_API_KEY = AUDIO_STT_OPENAI_API_KEY +app.state.config.STT_ENGINE = AUDIO_STT_ENGINE +app.state.config.STT_MODEL = AUDIO_STT_MODEL + +app.state.config.WHISPER_MODEL = WHISPER_MODEL + +app.state.config.TTS_OPENAI_API_BASE_URL = AUDIO_TTS_OPENAI_API_BASE_URL +app.state.config.TTS_OPENAI_API_KEY = AUDIO_TTS_OPENAI_API_KEY +app.state.config.TTS_ENGINE = AUDIO_TTS_ENGINE +app.state.config.TTS_MODEL = AUDIO_TTS_MODEL +app.state.config.TTS_VOICE = AUDIO_TTS_VOICE +app.state.config.TTS_API_KEY = AUDIO_TTS_API_KEY +app.state.config.TTS_SPLIT_ON = AUDIO_TTS_SPLIT_ON + + +app.state.config.TTS_AZURE_SPEECH_REGION = AUDIO_TTS_AZURE_SPEECH_REGION +app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT = AUDIO_TTS_AZURE_SPEECH_OUTPUT_FORMAT + + +app.state.faster_whisper_model = None +app.state.speech_synthesiser = None +app.state.speech_speaker_embeddings_dataset = None + + +######################################## +# +# TASKS +# +######################################## + app.state.config.TASK_MODEL = TASK_MODEL app.state.config.TASK_MODEL_EXTERNAL = TASK_MODEL_EXTERNAL -app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE = TITLE_GENERATION_PROMPT_TEMPLATE +app.state.config.ENABLE_SEARCH_QUERY_GENERATION = ENABLE_SEARCH_QUERY_GENERATION +app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION = ENABLE_RETRIEVAL_QUERY_GENERATION app.state.config.ENABLE_AUTOCOMPLETE_GENERATION = ENABLE_AUTOCOMPLETE_GENERATION +app.state.config.ENABLE_TAGS_GENERATION = ENABLE_TAGS_GENERATION + + +app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE = TITLE_GENERATION_PROMPT_TEMPLATE +app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE = TAGS_GENERATION_PROMPT_TEMPLATE +app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE = ( + TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE +) +app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE = QUERY_GENERATION_PROMPT_TEMPLATE +app.state.config.AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE = ( + AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE +) app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH = ( AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH ) -app.state.config.ENABLE_TAGS_GENERATION = ENABLE_TAGS_GENERATION -app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE = TAGS_GENERATION_PROMPT_TEMPLATE - -app.state.config.ENABLE_SEARCH_QUERY_GENERATION = ENABLE_SEARCH_QUERY_GENERATION -app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION = ENABLE_RETRIEVAL_QUERY_GENERATION -app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE = QUERY_GENERATION_PROMPT_TEMPLATE - -app.state.config.AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE = ( - AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE -) - -app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE = ( - TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE -) - -################################## +######################################## # -# ChatCompletion Middleware +# WEBUI # -################################## +######################################## - -def get_filter_function_ids(model): - def get_priority(function_id): - function = Functions.get_function_by_id(function_id) - if function is not None and hasattr(function, "valves"): - # TODO: Fix FunctionModel - return (function.valves if function.valves else {}).get("priority", 0) - return 0 - - filter_ids = [function.id for function in Functions.get_global_filter_functions()] - if "info" in model and "meta" in model["info"]: - filter_ids.extend(model["info"]["meta"].get("filterIds", [])) - filter_ids = list(set(filter_ids)) - - enabled_filter_ids = [ - function.id - for function in Functions.get_functions_by_type("filter", active_only=True) - ] - - filter_ids = [ - filter_id for filter_id in filter_ids if filter_id in enabled_filter_ids - ] - - filter_ids.sort(key=get_priority) - return filter_ids - - -async def chat_completion_filter_functions_handler(body, model, extra_params): - skip_files = None - - filter_ids = get_filter_function_ids(model) - for filter_id in filter_ids: - filter = Functions.get_function_by_id(filter_id) - if not filter: - continue - - if filter_id in webui_app.state.FUNCTIONS: - function_module = webui_app.state.FUNCTIONS[filter_id] - else: - function_module, _, _ = load_function_module_by_id(filter_id) - webui_app.state.FUNCTIONS[filter_id] = function_module - - # Check if the function has a file_handler variable - if hasattr(function_module, "file_handler"): - skip_files = function_module.file_handler - - if hasattr(function_module, "valves") and hasattr(function_module, "Valves"): - valves = Functions.get_function_valves_by_id(filter_id) - function_module.valves = function_module.Valves( - **(valves if valves else {}) - ) - - if not hasattr(function_module, "inlet"): - continue - - try: - inlet = function_module.inlet - - # Get the signature of the function - sig = inspect.signature(inlet) - params = {"body": body} | { - k: v - for k, v in { - **extra_params, - "__model__": model, - "__id__": filter_id, - }.items() - if k in sig.parameters - } - - if "__user__" in params and hasattr(function_module, "UserValves"): - try: - params["__user__"]["valves"] = function_module.UserValves( - **Functions.get_user_valves_by_id_and_user_id( - filter_id, params["__user__"]["id"] - ) - ) - except Exception as e: - print(e) - - if inspect.iscoroutinefunction(inlet): - body = await inlet(**params) - else: - body = inlet(**params) - - except Exception as e: - print(f"Error: {e}") - raise e - - if skip_files and "files" in body.get("metadata", {}): - del body["metadata"]["files"] - - return body, {} - - -def get_tools_function_calling_payload(messages, task_model_id, content): - user_message = get_last_user_message(messages) - history = "\n".join( - f"{message['role'].upper()}: \"\"\"{message['content']}\"\"\"" - for message in messages[::-1][:4] - ) - - prompt = f"History:\n{history}\nQuery: {user_message}" - - return { - "model": task_model_id, - "messages": [ - {"role": "system", "content": content}, - {"role": "user", "content": f"Query: {prompt}"}, - ], - "stream": False, - "metadata": {"task": str(TASKS.FUNCTION_CALLING)}, - } - - -async def get_content_from_response(response) -> Optional[str]: - content = None - if hasattr(response, "body_iterator"): - async for chunk in response.body_iterator: - data = json.loads(chunk.decode("utf-8")) - content = data["choices"][0]["message"]["content"] - - # Cleanup any remaining background tasks if necessary - if response.background is not None: - await response.background() - else: - content = response["choices"][0]["message"]["content"] - return content - - -def get_task_model_id( - default_model_id: str, task_model: str, task_model_external: str, models -) -> str: - # Set the task model - task_model_id = default_model_id - # Check if the user has a custom task model and use that model - if models[task_model_id]["owned_by"] == "ollama": - if task_model and task_model in models: - task_model_id = task_model - else: - if task_model_external and task_model_external in models: - task_model_id = task_model_external - - return task_model_id - - -async def chat_completion_tools_handler( - body: dict, user: UserModel, models, extra_params: dict -) -> tuple[dict, dict]: - # If tool_ids field is present, call the functions - metadata = body.get("metadata", {}) - - tool_ids = metadata.get("tool_ids", None) - log.debug(f"{tool_ids=}") - if not tool_ids: - return body, {} - - skip_files = False - sources = [] - - task_model_id = get_task_model_id( - body["model"], - app.state.config.TASK_MODEL, - app.state.config.TASK_MODEL_EXTERNAL, - models, - ) - tools = get_tools( - webui_app, - tool_ids, - user, - { - **extra_params, - "__model__": models[task_model_id], - "__messages__": body["messages"], - "__files__": metadata.get("files", []), - }, - ) - log.info(f"{tools=}") - - specs = [tool["spec"] for tool in tools.values()] - tools_specs = json.dumps(specs) - - if app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE != "": - template = app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE - else: - template = """Available Tools: {{TOOLS}}\nReturn an empty string if no tools match the query. If a function tool matches, construct and return a JSON object in the format {\"name\": \"functionName\", \"parameters\": {\"requiredFunctionParamKey\": \"requiredFunctionParamValue\"}} using the appropriate tool and its parameters. Only return the object and limit the response to the JSON object without additional text.""" - - tools_function_calling_prompt = tools_function_calling_generation_template( - template, tools_specs - ) - log.info(f"{tools_function_calling_prompt=}") - payload = get_tools_function_calling_payload( - body["messages"], task_model_id, tools_function_calling_prompt - ) - - try: - payload = filter_pipeline(payload, user, models) - except Exception as e: - raise e - - try: - response = await generate_chat_completions(form_data=payload, user=user) - log.debug(f"{response=}") - content = await get_content_from_response(response) - log.debug(f"{content=}") - - if not content: - return body, {} - - try: - content = content[content.find("{") : content.rfind("}") + 1] - if not content: - raise Exception("No JSON object found in the response") - - result = json.loads(content) - - tool_function_name = result.get("name", None) - if tool_function_name not in tools: - return body, {} - - tool_function_params = result.get("parameters", {}) - - try: - required_params = ( - tools[tool_function_name] - .get("spec", {}) - .get("parameters", {}) - .get("required", []) - ) - tool_function = tools[tool_function_name]["callable"] - tool_function_params = { - k: v - for k, v in tool_function_params.items() - if k in required_params - } - tool_output = await tool_function(**tool_function_params) - - except Exception as e: - tool_output = str(e) - - if isinstance(tool_output, str): - if tools[tool_function_name]["citation"]: - sources.append( - { - "source": { - "name": f"TOOL:{tools[tool_function_name]['toolkit_id']}/{tool_function_name}" - }, - "document": [tool_output], - "metadata": [ - { - "source": f"TOOL:{tools[tool_function_name]['toolkit_id']}/{tool_function_name}" - } - ], - } - ) - else: - sources.append( - { - "source": {}, - "document": [tool_output], - "metadata": [ - { - "source": f"TOOL:{tools[tool_function_name]['toolkit_id']}/{tool_function_name}" - } - ], - } - ) - - if tools[tool_function_name]["file_handler"]: - skip_files = True - - except Exception as e: - log.exception(f"Error: {e}") - content = None - except Exception as e: - log.exception(f"Error: {e}") - content = None - - log.debug(f"tool_contexts: {sources}") - - if skip_files and "files" in body.get("metadata", {}): - del body["metadata"]["files"] - - return body, {"sources": sources} - - -async def chat_completion_files_handler( - body: dict, user: UserModel -) -> tuple[dict, dict[str, list]]: - sources = [] - - if files := body.get("metadata", {}).get("files", None): - try: - queries_response = await generate_queries( - { - "model": body["model"], - "messages": body["messages"], - "type": "retrieval", - }, - user, - ) - queries_response = queries_response["choices"][0]["message"]["content"] - - try: - bracket_start = queries_response.find("{") - bracket_end = queries_response.rfind("}") + 1 - - if bracket_start == -1 or bracket_end == -1: - raise Exception("No JSON object found in the response") - - queries_response = queries_response[bracket_start:bracket_end] - queries_response = json.loads(queries_response) - except Exception as e: - queries_response = {"queries": [queries_response]} - - queries = queries_response.get("queries", []) - except Exception as e: - queries = [] - - if len(queries) == 0: - queries = [get_last_user_message(body["messages"])] - - sources = get_sources_from_files( - files=files, - queries=queries, - embedding_function=retrieval_app.state.EMBEDDING_FUNCTION, - k=retrieval_app.state.config.TOP_K, - reranking_function=retrieval_app.state.sentence_transformer_rf, - r=retrieval_app.state.config.RELEVANCE_THRESHOLD, - hybrid_search=retrieval_app.state.config.ENABLE_RAG_HYBRID_SEARCH, - ) - - log.debug(f"rag_contexts:sources: {sources}") - return body, {"sources": sources} - - -def is_chat_completion_request(request): - return request.method == "POST" and any( - endpoint in request.url.path - for endpoint in ["/ollama/api/chat", "/chat/completions"] - ) - - -async def get_body_and_model_and_user(request, models): - # Read the original request body - body = await request.body() - body_str = body.decode("utf-8") - body = json.loads(body_str) if body_str else {} - - model_id = body["model"] - if model_id not in models: - raise Exception("Model not found") - model = models[model_id] - - user = get_current_user( - request, - get_http_authorization_cred(request.headers.get("Authorization")), - ) - - return body, model, user - - -class ChatCompletionMiddleware(BaseHTTPMiddleware): - async def dispatch(self, request: Request, call_next): - if not is_chat_completion_request(request): - return await call_next(request) - log.debug(f"request.url.path: {request.url.path}") - - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} - - try: - body, model, user = await get_body_and_model_and_user(request, models) - except Exception as e: - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - - model_info = Models.get_model_by_id(model["id"]) - if user.role == "user" and not BYPASS_MODEL_ACCESS_CONTROL: - if model.get("arena"): - if not has_access( - user.id, - type="read", - access_control=model.get("info", {}) - .get("meta", {}) - .get("access_control", {}), - ): - raise HTTPException( - status_code=403, - detail="Model not found", - ) - else: - if not model_info: - return JSONResponse( - status_code=status.HTTP_404_NOT_FOUND, - content={"detail": "Model not found"}, - ) - elif not ( - user.id == model_info.user_id - or has_access( - user.id, type="read", access_control=model_info.access_control - ) - ): - return JSONResponse( - status_code=status.HTTP_403_FORBIDDEN, - content={"detail": "User does not have access to the model"}, - ) - - metadata = { - "chat_id": body.pop("chat_id", None), - "message_id": body.pop("id", None), - "session_id": body.pop("session_id", None), - "tool_ids": body.get("tool_ids", None), - "files": body.get("files", None), - } - body["metadata"] = metadata - - extra_params = { - "__event_emitter__": get_event_emitter(metadata), - "__event_call__": get_event_call(metadata), - "__user__": { - "id": user.id, - "email": user.email, - "name": user.name, - "role": user.role, - }, - "__metadata__": metadata, - } - - # Initialize data_items to store additional data to be sent to the client - # Initialize contexts and citation - data_items = [] - sources = [] - - try: - body, flags = await chat_completion_filter_functions_handler( - body, model, extra_params - ) - except Exception as e: - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - - tool_ids = body.pop("tool_ids", None) - files = body.pop("files", None) - - metadata = { - **metadata, - "tool_ids": tool_ids, - "files": files, - } - body["metadata"] = metadata - - try: - body, flags = await chat_completion_tools_handler( - body, user, models, extra_params - ) - sources.extend(flags.get("sources", [])) - except Exception as e: - log.exception(e) - - try: - body, flags = await chat_completion_files_handler(body, user) - sources.extend(flags.get("sources", [])) - except Exception as e: - log.exception(e) - - # If context is not empty, insert it into the messages - if len(sources) > 0: - context_string = "" - for source_idx, source in enumerate(sources): - source_id = source.get("source", {}).get("name", "") - - if "document" in source: - for doc_idx, doc_context in enumerate(source["document"]): - metadata = source.get("metadata") - doc_source_id = None - - if metadata: - doc_source_id = metadata[doc_idx].get("source", source_id) - - if source_id: - context_string += f"{doc_source_id if doc_source_id is not None else source_id}{doc_context}\n" - else: - # If there is no source_id, then do not include the source_id tag - context_string += f"{doc_context}\n" - - context_string = context_string.strip() - prompt = get_last_user_message(body["messages"]) - - if prompt is None: - raise Exception("No user message found") - if ( - retrieval_app.state.config.RELEVANCE_THRESHOLD == 0 - and context_string.strip() == "" - ): - log.debug( - f"With a 0 relevancy threshold for RAG, the context cannot be empty" - ) - - # Workaround for Ollama 2.0+ system prompt issue - # TODO: replace with add_or_update_system_message - if model["owned_by"] == "ollama": - body["messages"] = prepend_to_first_user_message_content( - rag_template( - retrieval_app.state.config.RAG_TEMPLATE, context_string, prompt - ), - body["messages"], - ) - else: - body["messages"] = add_or_update_system_message( - rag_template( - retrieval_app.state.config.RAG_TEMPLATE, context_string, prompt - ), - body["messages"], - ) - - # If there are citations, add them to the data_items - sources = [ - source for source in sources if source.get("source", {}).get("name", "") - ] - if len(sources) > 0: - data_items.append({"sources": sources}) - - modified_body_bytes = json.dumps(body).encode("utf-8") - # Replace the request body with the modified one - request._body = modified_body_bytes - # Set custom header to ensure content-length matches new body length - request.headers.__dict__["_list"] = [ - (b"content-length", str(len(modified_body_bytes)).encode("utf-8")), - *[(k, v) for k, v in request.headers.raw if k.lower() != b"content-length"], - ] - - response = await call_next(request) - if not isinstance(response, StreamingResponse): - return response - - content_type = response.headers["Content-Type"] - is_openai = "text/event-stream" in content_type - is_ollama = "application/x-ndjson" in content_type - if not is_openai and not is_ollama: - return response - - def wrap_item(item): - return f"data: {item}\n\n" if is_openai else f"{item}\n" - - async def stream_wrapper(original_generator, data_items): - for item in data_items: - yield wrap_item(json.dumps(item)) - - async for data in original_generator: - yield data - - return StreamingResponse( - stream_wrapper(response.body_iterator, data_items), - headers=dict(response.headers), - ) - - async def _receive(self, body: bytes): - return {"type": "http.request", "body": body, "more_body": False} - - -app.add_middleware(ChatCompletionMiddleware) - - -################################## -# -# Pipeline Middleware -# -################################## - - -def get_sorted_filters(model_id, models): - filters = [ - model - for model in models.values() - if "pipeline" in model - and "type" in model["pipeline"] - and model["pipeline"]["type"] == "filter" - and ( - model["pipeline"]["pipelines"] == ["*"] - or any( - model_id == target_model_id - for target_model_id in model["pipeline"]["pipelines"] - ) - ) - ] - sorted_filters = sorted(filters, key=lambda x: x["pipeline"]["priority"]) - return sorted_filters - - -def filter_pipeline(payload, user, models): - user = {"id": user.id, "email": user.email, "name": user.name, "role": user.role} - model_id = payload["model"] - - sorted_filters = get_sorted_filters(model_id, models) - model = models[model_id] - - if "pipeline" in model: - sorted_filters.append(model) - - for filter in sorted_filters: - r = None - try: - urlIdx = filter["urlIdx"] - - url = openai_app.state.config.OPENAI_API_BASE_URLS[urlIdx] - key = openai_app.state.config.OPENAI_API_KEYS[urlIdx] - - if key == "": - continue - - headers = {"Authorization": f"Bearer {key}"} - r = requests.post( - f"{url}/{filter['id']}/filter/inlet", - headers=headers, - json={ - "user": user, - "body": payload, - }, - ) - - r.raise_for_status() - payload = r.json() - except Exception as e: - # Handle connection error here - print(f"Connection error: {e}") - - if r is not None: - res = r.json() - if "detail" in res: - raise Exception(r.status_code, res["detail"]) - - return payload - - -class PipelineMiddleware(BaseHTTPMiddleware): - async def dispatch(self, request: Request, call_next): - if not is_chat_completion_request(request): - return await call_next(request) - - log.debug(f"request.url.path: {request.url.path}") - - # Read the original request body - body = await request.body() - # Decode body to string - body_str = body.decode("utf-8") - # Parse string to JSON - data = json.loads(body_str) if body_str else {} - - try: - user = get_current_user( - request, - get_http_authorization_cred(request.headers["Authorization"]), - ) - except KeyError as e: - if len(e.args) > 1: - return JSONResponse( - status_code=e.args[0], - content={"detail": e.args[1]}, - ) - else: - return JSONResponse( - status_code=status.HTTP_401_UNAUTHORIZED, - content={"detail": "Not authenticated"}, - ) - except HTTPException as e: - return JSONResponse( - status_code=e.status_code, - content={"detail": e.detail}, - ) - - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} - - try: - data = filter_pipeline(data, user, models) - except Exception as e: - if len(e.args) > 1: - return JSONResponse( - status_code=e.args[0], - content={"detail": e.args[1]}, - ) - else: - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - - modified_body_bytes = json.dumps(data).encode("utf-8") - # Replace the request body with the modified one - request._body = modified_body_bytes - # Set custom header to ensure content-length matches new body length - request.headers.__dict__["_list"] = [ - (b"content-length", str(len(modified_body_bytes)).encode("utf-8")), - *[(k, v) for k, v in request.headers.raw if k.lower() != b"content-length"], - ] - - response = await call_next(request) - return response - - async def _receive(self, body: bytes): - return {"type": "http.request", "body": body, "more_body": False} - - -app.add_middleware(PipelineMiddleware) - - -from urllib.parse import urlencode, parse_qs, urlparse +app.state.MODELS = {} class RedirectMiddleware(BaseHTTPMiddleware): @@ -975,16 +656,6 @@ class RedirectMiddleware(BaseHTTPMiddleware): # Add the middleware to the app app.add_middleware(RedirectMiddleware) - - -app.add_middleware( - CORSMiddleware, - allow_origins=CORS_ALLOW_ORIGIN, - allow_credentials=True, - allow_methods=["*"], - allow_headers=["*"], -) - app.add_middleware(SecurityHeadersMiddleware) @@ -999,21 +670,13 @@ async def commit_session_after_request(request: Request, call_next): @app.middleware("http") async def check_url(request: Request, call_next): start_time = int(time.time()) - request.state.enable_api_key = webui_app.state.config.ENABLE_API_KEY + request.state.enable_api_key = app.state.config.ENABLE_API_KEY response = await call_next(request) process_time = int(time.time()) - start_time response.headers["X-Process-Time"] = str(process_time) return response -@app.middleware("http") -async def update_embedding_function(request: Request, call_next): - response = await call_next(request) - if "/embedding/update" in request.url.path: - webui_app.state.EMBEDDING_FUNCTION = retrieval_app.state.EMBEDDING_FUNCTION - return response - - @app.middleware("http") async def inspect_websocket(request: Request, call_next): if ( @@ -1032,198 +695,61 @@ async def inspect_websocket(request: Request, call_next): return await call_next(request) +app.add_middleware( + CORSMiddleware, + allow_origins=CORS_ALLOW_ORIGIN, + allow_credentials=True, + allow_methods=["*"], + allow_headers=["*"], +) + + app.mount("/ws", socket_app) -app.mount("/ollama", ollama_app) -app.mount("/openai", openai_app) - -app.mount("/images/api/v1", images_app) -app.mount("/audio/api/v1", audio_app) -app.mount("/retrieval/api/v1", retrieval_app) - -app.mount("/api/v1", webui_app) - -webui_app.state.EMBEDDING_FUNCTION = retrieval_app.state.EMBEDDING_FUNCTION -async def get_all_base_models(): - open_webui_models = [] - openai_models = [] - ollama_models = [] - - if app.state.config.ENABLE_OPENAI_API: - openai_models = await get_openai_models() - openai_models = openai_models["data"] - - if app.state.config.ENABLE_OLLAMA_API: - ollama_models = await get_ollama_models() - ollama_models = [ - { - "id": model["model"], - "name": model["name"], - "object": "model", - "created": int(time.time()), - "owned_by": "ollama", - "ollama": model, - } - for model in ollama_models["models"] - ] - - open_webui_models = await get_open_webui_models() - - models = open_webui_models + openai_models + ollama_models - return models +app.include_router(ollama.router, prefix="/ollama", tags=["ollama"]) +app.include_router(openai.router, prefix="/openai", tags=["openai"]) -@cached(ttl=3) -async def get_all_models(): - models = await get_all_base_models() +app.include_router(pipelines.router, prefix="/api/v1/pipelines", tags=["pipelines"]) +app.include_router(tasks.router, prefix="/api/v1/tasks", tags=["tasks"]) +app.include_router(images.router, prefix="/api/v1/images", tags=["images"]) +app.include_router(audio.router, prefix="/api/v1/audio", tags=["audio"]) +app.include_router(retrieval.router, prefix="/api/v1/retrieval", tags=["retrieval"]) - # If there are no models, return an empty list - if len([model for model in models if not model.get("arena", False)]) == 0: - return [] +app.include_router(configs.router, prefix="/api/v1/configs", tags=["configs"]) - global_action_ids = [ - function.id for function in Functions.get_global_action_functions() - ] - enabled_action_ids = [ - function.id - for function in Functions.get_functions_by_type("action", active_only=True) - ] +app.include_router(auths.router, prefix="/api/v1/auths", tags=["auths"]) +app.include_router(users.router, prefix="/api/v1/users", tags=["users"]) - custom_models = Models.get_all_models() - for custom_model in custom_models: - if custom_model.base_model_id is None: - for model in models: - if ( - custom_model.id == model["id"] - or custom_model.id == model["id"].split(":")[0] - ): - if custom_model.is_active: - model["name"] = custom_model.name - model["info"] = custom_model.model_dump() +app.include_router(chats.router, prefix="/api/v1/chats", tags=["chats"]) - action_ids = [] - if "info" in model and "meta" in model["info"]: - action_ids.extend( - model["info"]["meta"].get("actionIds", []) - ) +app.include_router(models.router, prefix="/api/v1/models", tags=["models"]) +app.include_router(knowledge.router, prefix="/api/v1/knowledge", tags=["knowledge"]) +app.include_router(prompts.router, prefix="/api/v1/prompts", tags=["prompts"]) +app.include_router(tools.router, prefix="/api/v1/tools", tags=["tools"]) - model["action_ids"] = action_ids - else: - models.remove(model) +app.include_router(memories.router, prefix="/api/v1/memories", tags=["memories"]) +app.include_router(folders.router, prefix="/api/v1/folders", tags=["folders"]) +app.include_router(groups.router, prefix="/api/v1/groups", tags=["groups"]) +app.include_router(files.router, prefix="/api/v1/files", tags=["files"]) +app.include_router(functions.router, prefix="/api/v1/functions", tags=["functions"]) +app.include_router( + evaluations.router, prefix="/api/v1/evaluations", tags=["evaluations"] +) +app.include_router(utils.router, prefix="/api/v1/utils", tags=["utils"]) - elif custom_model.is_active and ( - custom_model.id not in [model["id"] for model in models] - ): - owned_by = "openai" - pipe = None - action_ids = [] - for model in models: - if ( - custom_model.base_model_id == model["id"] - or custom_model.base_model_id == model["id"].split(":")[0] - ): - owned_by = model["owned_by"] - if "pipe" in model: - pipe = model["pipe"] - break - - if custom_model.meta: - meta = custom_model.meta.model_dump() - if "actionIds" in meta: - action_ids.extend(meta["actionIds"]) - - models.append( - { - "id": f"{custom_model.id}", - "name": custom_model.name, - "object": "model", - "created": custom_model.created_at, - "owned_by": owned_by, - "info": custom_model.model_dump(), - "preset": True, - **({"pipe": pipe} if pipe is not None else {}), - "action_ids": action_ids, - } - ) - - # Process action_ids to get the actions - def get_action_items_from_module(function, module): - actions = [] - if hasattr(module, "actions"): - actions = module.actions - return [ - { - "id": f"{function.id}.{action['id']}", - "name": action.get("name", f"{function.name} ({action['id']})"), - "description": function.meta.description, - "icon_url": action.get( - "icon_url", function.meta.manifest.get("icon_url", None) - ), - } - for action in actions - ] - else: - return [ - { - "id": function.id, - "name": function.name, - "description": function.meta.description, - "icon_url": function.meta.manifest.get("icon_url", None), - } - ] - - def get_function_module_by_id(function_id): - if function_id in webui_app.state.FUNCTIONS: - function_module = webui_app.state.FUNCTIONS[function_id] - else: - function_module, _, _ = load_function_module_by_id(function_id) - webui_app.state.FUNCTIONS[function_id] = function_module - - for model in models: - action_ids = [ - action_id - for action_id in list(set(model.pop("action_ids", []) + global_action_ids)) - if action_id in enabled_action_ids - ] - - model["actions"] = [] - for action_id in action_ids: - action_function = Functions.get_function_by_id(action_id) - if action_function is None: - raise Exception(f"Action not found: {action_id}") - - function_module = get_function_module_by_id(action_id) - model["actions"].extend( - get_action_items_from_module(action_function, function_module) - ) - log.debug(f"get_all_models() returned {len(models)} models") - - return models +################################## +# +# Chat Endpoints +# +################################## @app.get("/api/models") -async def get_models(user=Depends(get_verified_user)): - models = await get_all_models() - - # Filter out filter pipelines - models = [ - model - for model in models - if "pipeline" not in model or model["pipeline"].get("type", None) != "filter" - ] - - model_order_list = webui_app.state.config.MODEL_ORDER_LIST - if model_order_list: - model_order_dict = {model_id: i for i, model_id in enumerate(model_order_list)} - # Sort models by order list priority, with fallback for those not in the list - models.sort( - key=lambda x: (model_order_dict.get(x["id"], float("inf")), x["name"]) - ) - - # Filter out models that the user does not have access to - if user.role == "user" and not BYPASS_MODEL_ACCESS_CONTROL: +async def get_models(request: Request, user=Depends(get_verified_user)): + def get_filtered_models(models, user): filtered_models = [] for model in models: if model.get("arena"): @@ -1243,1320 +769,112 @@ async def get_models(user=Depends(get_verified_user)): user.id, type="read", access_control=model_info.access_control ): filtered_models.append(model) - models = filtered_models + + return filtered_models + + models = await get_all_models(request) + + # Filter out filter pipelines + models = [ + model + for model in models + if "pipeline" not in model or model["pipeline"].get("type", None) != "filter" + ] + + model_order_list = request.app.state.config.MODEL_ORDER_LIST + if model_order_list: + model_order_dict = {model_id: i for i, model_id in enumerate(model_order_list)} + # Sort models by order list priority, with fallback for those not in the list + models.sort( + key=lambda x: (model_order_dict.get(x["id"], float("inf")), x["name"]) + ) + + # Filter out models that the user does not have access to + if user.role == "user" and not BYPASS_MODEL_ACCESS_CONTROL: + models = get_filtered_models(models, user) log.debug( f"/api/models returned filtered models accessible to the user: {json.dumps([model['id'] for model in models])}" ) - return {"data": models} @app.get("/api/models/base") -async def get_base_models(user=Depends(get_admin_user)): - models = await get_all_base_models() - - # Filter out arena models - models = [model for model in models if not model.get("arena", False)] +async def get_base_models(request: Request, user=Depends(get_admin_user)): + models = await get_all_base_models(request) return {"data": models} @app.post("/api/chat/completions") -async def generate_chat_completions( - form_data: dict, user=Depends(get_verified_user), bypass_filter: bool = False +async def chat_completion( + request: Request, + form_data: dict, + user=Depends(get_verified_user), + bypass_filter: bool = False, ): - if BYPASS_MODEL_ACCESS_CONTROL: - bypass_filter = True + if not request.app.state.MODELS: + await get_all_models(request) - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} + try: + model_id = form_data.get("model", None) + if model_id not in request.app.state.MODELS: + raise Exception("Model not found") + model = request.app.state.MODELS[model_id] - model_id = form_data["model"] - if model_id not in models: + # Check if user has access to the model + if not bypass_filter and user.role == "user": + try: + check_model_access(user, model) + except Exception as e: + raise e + + form_data, events = await process_chat_payload(request, form_data, user, model) + except Exception as e: raise HTTPException( - status_code=status.HTTP_404_NOT_FOUND, - detail="Model not found", + status_code=status.HTTP_400_BAD_REQUEST, + detail=str(e), ) - model = models[model_id] - - # Check if user has access to the model - if not bypass_filter and user.role == "user": - if model.get("arena"): - if not has_access( - user.id, - type="read", - access_control=model.get("info", {}) - .get("meta", {}) - .get("access_control", {}), - ): - raise HTTPException( - status_code=403, - detail="Model not found", - ) - else: - model_info = Models.get_model_by_id(model_id) - if not model_info: - raise HTTPException( - status_code=404, - detail="Model not found", - ) - elif not ( - user.id == model_info.user_id - or has_access( - user.id, type="read", access_control=model_info.access_control - ) - ): - raise HTTPException( - status_code=403, - detail="Model not found", - ) - - if model["owned_by"] == "arena": - model_ids = model.get("info", {}).get("meta", {}).get("model_ids") - filter_mode = model.get("info", {}).get("meta", {}).get("filter_mode") - if model_ids and filter_mode == "exclude": - model_ids = [ - model["id"] - for model in await get_all_models() - if model.get("owned_by") != "arena" and model["id"] not in model_ids - ] - - selected_model_id = None - if isinstance(model_ids, list) and model_ids: - selected_model_id = random.choice(model_ids) - else: - model_ids = [ - model["id"] - for model in await get_all_models() - if model.get("owned_by") != "arena" - ] - selected_model_id = random.choice(model_ids) - - form_data["model"] = selected_model_id - - if form_data.get("stream") == True: - - async def stream_wrapper(stream): - yield f"data: {json.dumps({'selected_model_id': selected_model_id})}\n\n" - async for chunk in stream: - yield chunk - - response = await generate_chat_completions( - form_data, user, bypass_filter=True - ) - return StreamingResponse( - stream_wrapper(response.body_iterator), media_type="text/event-stream" - ) - else: - return { - **( - await generate_chat_completions(form_data, user, bypass_filter=True) - ), - "selected_model_id": selected_model_id, - } - - if model.get("pipe"): - # Below does not require bypass_filter because this is the only route the uses this function and it is already bypassing the filter - return await generate_function_chat_completion( - form_data, user=user, models=models + try: + response = await chat_completion_handler( + request, form_data, user, bypass_filter ) - if model["owned_by"] == "ollama": - # Using /ollama/api/chat endpoint - form_data = convert_payload_openai_to_ollama(form_data) - form_data = GenerateChatCompletionForm(**form_data) - response = await generate_ollama_chat_completion( - form_data=form_data, user=user, bypass_filter=bypass_filter - ) - if form_data.stream: - response.headers["content-type"] = "text/event-stream" - return StreamingResponse( - convert_streaming_response_ollama_to_openai(response), - headers=dict(response.headers), - ) - else: - return convert_response_ollama_to_openai(response) - else: - return await generate_openai_chat_completion( - form_data, user=user, bypass_filter=bypass_filter + return await process_chat_response(response, events) + except Exception as e: + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=str(e), ) +# Alias for chat_completion (Legacy) +generate_chat_completions = chat_completion +generate_chat_completion = chat_completion + + @app.post("/api/chat/completed") -async def chat_completed(form_data: dict, user=Depends(get_verified_user)): - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} - - data = form_data - model_id = data["model"] - if model_id not in models: +async def chat_completed( + request: Request, form_data: dict, user=Depends(get_verified_user) +): + try: + return await chat_completed_handler(request, form_data, user) + except Exception as e: raise HTTPException( - status_code=status.HTTP_404_NOT_FOUND, - detail="Model not found", + status_code=status.HTTP_400_BAD_REQUEST, + detail=str(e), ) - model = models[model_id] - sorted_filters = get_sorted_filters(model_id, models) - if "pipeline" in model: - sorted_filters = [model] + sorted_filters - - for filter in sorted_filters: - r = None - try: - urlIdx = filter["urlIdx"] - - url = openai_app.state.config.OPENAI_API_BASE_URLS[urlIdx] - key = openai_app.state.config.OPENAI_API_KEYS[urlIdx] - - if key != "": - headers = {"Authorization": f"Bearer {key}"} - r = requests.post( - f"{url}/{filter['id']}/filter/outlet", - headers=headers, - json={ - "user": { - "id": user.id, - "name": user.name, - "email": user.email, - "role": user.role, - }, - "body": data, - }, - ) - - r.raise_for_status() - data = r.json() - except Exception as e: - # Handle connection error here - print(f"Connection error: {e}") - - if r is not None: - try: - res = r.json() - if "detail" in res: - return JSONResponse( - status_code=r.status_code, - content=res, - ) - except Exception: - pass - - else: - pass - - __event_emitter__ = get_event_emitter( - { - "chat_id": data["chat_id"], - "message_id": data["id"], - "session_id": data["session_id"], - } - ) - - __event_call__ = get_event_call( - { - "chat_id": data["chat_id"], - "message_id": data["id"], - "session_id": data["session_id"], - } - ) - - def get_priority(function_id): - function = Functions.get_function_by_id(function_id) - if function is not None and hasattr(function, "valves"): - # TODO: Fix FunctionModel to include vavles - return (function.valves if function.valves else {}).get("priority", 0) - return 0 - - filter_ids = [function.id for function in Functions.get_global_filter_functions()] - if "info" in model and "meta" in model["info"]: - filter_ids.extend(model["info"]["meta"].get("filterIds", [])) - filter_ids = list(set(filter_ids)) - - enabled_filter_ids = [ - function.id - for function in Functions.get_functions_by_type("filter", active_only=True) - ] - filter_ids = [ - filter_id for filter_id in filter_ids if filter_id in enabled_filter_ids - ] - - # Sort filter_ids by priority, using the get_priority function - filter_ids.sort(key=get_priority) - - for filter_id in filter_ids: - filter = Functions.get_function_by_id(filter_id) - if not filter: - continue - - if filter_id in webui_app.state.FUNCTIONS: - function_module = webui_app.state.FUNCTIONS[filter_id] - else: - function_module, _, _ = load_function_module_by_id(filter_id) - webui_app.state.FUNCTIONS[filter_id] = function_module - - if hasattr(function_module, "valves") and hasattr(function_module, "Valves"): - valves = Functions.get_function_valves_by_id(filter_id) - function_module.valves = function_module.Valves( - **(valves if valves else {}) - ) - - if not hasattr(function_module, "outlet"): - continue - try: - outlet = function_module.outlet - - # Get the signature of the function - sig = inspect.signature(outlet) - params = {"body": data} - - # Extra parameters to be passed to the function - extra_params = { - "__model__": model, - "__id__": filter_id, - "__event_emitter__": __event_emitter__, - "__event_call__": __event_call__, - } - - # Add extra params in contained in function signature - for key, value in extra_params.items(): - if key in sig.parameters: - params[key] = value - - if "__user__" in sig.parameters: - __user__ = { - "id": user.id, - "email": user.email, - "name": user.name, - "role": user.role, - } - - try: - if hasattr(function_module, "UserValves"): - __user__["valves"] = function_module.UserValves( - **Functions.get_user_valves_by_id_and_user_id( - filter_id, user.id - ) - ) - except Exception as e: - print(e) - - params = {**params, "__user__": __user__} - - if inspect.iscoroutinefunction(outlet): - data = await outlet(**params) - else: - data = outlet(**params) - - except Exception as e: - print(f"Error: {e}") - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - - return data - @app.post("/api/chat/actions/{action_id}") -async def chat_action(action_id: str, form_data: dict, user=Depends(get_verified_user)): - if "." in action_id: - action_id, sub_action_id = action_id.split(".") - else: - sub_action_id = None - - action = Functions.get_function_by_id(action_id) - if not action: - raise HTTPException( - status_code=status.HTTP_404_NOT_FOUND, - detail="Action not found", - ) - - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} - - data = form_data - model_id = data["model"] - - if model_id not in models: - raise HTTPException( - status_code=status.HTTP_404_NOT_FOUND, - detail="Model not found", - ) - model = models[model_id] - - __event_emitter__ = get_event_emitter( - { - "chat_id": data["chat_id"], - "message_id": data["id"], - "session_id": data["session_id"], - } - ) - __event_call__ = get_event_call( - { - "chat_id": data["chat_id"], - "message_id": data["id"], - "session_id": data["session_id"], - } - ) - - if action_id in webui_app.state.FUNCTIONS: - function_module = webui_app.state.FUNCTIONS[action_id] - else: - function_module, _, _ = load_function_module_by_id(action_id) - webui_app.state.FUNCTIONS[action_id] = function_module - - if hasattr(function_module, "valves") and hasattr(function_module, "Valves"): - valves = Functions.get_function_valves_by_id(action_id) - function_module.valves = function_module.Valves(**(valves if valves else {})) - - if hasattr(function_module, "action"): - try: - action = function_module.action - - # Get the signature of the function - sig = inspect.signature(action) - params = {"body": data} - - # Extra parameters to be passed to the function - extra_params = { - "__model__": model, - "__id__": sub_action_id if sub_action_id is not None else action_id, - "__event_emitter__": __event_emitter__, - "__event_call__": __event_call__, - } - - # Add extra params in contained in function signature - for key, value in extra_params.items(): - if key in sig.parameters: - params[key] = value - - if "__user__" in sig.parameters: - __user__ = { - "id": user.id, - "email": user.email, - "name": user.name, - "role": user.role, - } - - try: - if hasattr(function_module, "UserValves"): - __user__["valves"] = function_module.UserValves( - **Functions.get_user_valves_by_id_and_user_id( - action_id, user.id - ) - ) - except Exception as e: - print(e) - - params = {**params, "__user__": __user__} - - if inspect.iscoroutinefunction(action): - data = await action(**params) - else: - data = action(**params) - - except Exception as e: - print(f"Error: {e}") - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - - return data - - -################################## -# -# Task Endpoints -# -################################## - - -# TODO: Refactor task API endpoints below into a separate file - - -@app.get("/api/task/config") -async def get_task_config(user=Depends(get_verified_user)): - return { - "TASK_MODEL": app.state.config.TASK_MODEL, - "TASK_MODEL_EXTERNAL": app.state.config.TASK_MODEL_EXTERNAL, - "TITLE_GENERATION_PROMPT_TEMPLATE": app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE, - "ENABLE_AUTOCOMPLETE_GENERATION": app.state.config.ENABLE_AUTOCOMPLETE_GENERATION, - "AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH": app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH, - "TAGS_GENERATION_PROMPT_TEMPLATE": app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE, - "ENABLE_TAGS_GENERATION": app.state.config.ENABLE_TAGS_GENERATION, - "ENABLE_SEARCH_QUERY_GENERATION": app.state.config.ENABLE_SEARCH_QUERY_GENERATION, - "ENABLE_RETRIEVAL_QUERY_GENERATION": app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION, - "QUERY_GENERATION_PROMPT_TEMPLATE": app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE, - "TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE": app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE, - } - - -class TaskConfigForm(BaseModel): - TASK_MODEL: Optional[str] - TASK_MODEL_EXTERNAL: Optional[str] - TITLE_GENERATION_PROMPT_TEMPLATE: str - ENABLE_AUTOCOMPLETE_GENERATION: bool - AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH: int - TAGS_GENERATION_PROMPT_TEMPLATE: str - ENABLE_TAGS_GENERATION: bool - ENABLE_SEARCH_QUERY_GENERATION: bool - ENABLE_RETRIEVAL_QUERY_GENERATION: bool - QUERY_GENERATION_PROMPT_TEMPLATE: str - TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE: str - - -@app.post("/api/task/config/update") -async def update_task_config(form_data: TaskConfigForm, user=Depends(get_admin_user)): - app.state.config.TASK_MODEL = form_data.TASK_MODEL - app.state.config.TASK_MODEL_EXTERNAL = form_data.TASK_MODEL_EXTERNAL - app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE = ( - form_data.TITLE_GENERATION_PROMPT_TEMPLATE - ) - - app.state.config.ENABLE_AUTOCOMPLETE_GENERATION = ( - form_data.ENABLE_AUTOCOMPLETE_GENERATION - ) - app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH = ( - form_data.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH - ) - - app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE = ( - form_data.TAGS_GENERATION_PROMPT_TEMPLATE - ) - app.state.config.ENABLE_TAGS_GENERATION = form_data.ENABLE_TAGS_GENERATION - app.state.config.ENABLE_SEARCH_QUERY_GENERATION = ( - form_data.ENABLE_SEARCH_QUERY_GENERATION - ) - app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION = ( - form_data.ENABLE_RETRIEVAL_QUERY_GENERATION - ) - - app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE = ( - form_data.QUERY_GENERATION_PROMPT_TEMPLATE - ) - app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE = ( - form_data.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE - ) - - return { - "TASK_MODEL": app.state.config.TASK_MODEL, - "TASK_MODEL_EXTERNAL": app.state.config.TASK_MODEL_EXTERNAL, - "TITLE_GENERATION_PROMPT_TEMPLATE": app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE, - "ENABLE_AUTOCOMPLETE_GENERATION": app.state.config.ENABLE_AUTOCOMPLETE_GENERATION, - "AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH": app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH, - "TAGS_GENERATION_PROMPT_TEMPLATE": app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE, - "ENABLE_TAGS_GENERATION": app.state.config.ENABLE_TAGS_GENERATION, - "ENABLE_SEARCH_QUERY_GENERATION": app.state.config.ENABLE_SEARCH_QUERY_GENERATION, - "ENABLE_RETRIEVAL_QUERY_GENERATION": app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION, - "QUERY_GENERATION_PROMPT_TEMPLATE": app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE, - "TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE": app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE, - } - - -@app.post("/api/task/title/completions") -async def generate_title(form_data: dict, user=Depends(get_verified_user)): - - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} - - model_id = form_data["model"] - if model_id not in models: - raise HTTPException( - status_code=status.HTTP_404_NOT_FOUND, - detail="Model not found", - ) - - # Check if the user has a custom task model - # If the user has a custom task model, use that model - task_model_id = get_task_model_id( - model_id, - app.state.config.TASK_MODEL, - app.state.config.TASK_MODEL_EXTERNAL, - models, - ) - - log.debug( - f"generating chat title using model {task_model_id} for user {user.email} " - ) - - if app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE != "": - template = app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE - else: - template = """Create a concise, 3-5 word title with an emoji as a title for the chat history, in the given language. Suitable Emojis for the summary can be used to enhance understanding but avoid quotation marks or special formatting. RESPOND ONLY WITH THE TITLE TEXT. - -Examples of titles: -📉 Stock Market Trends -🍪 Perfect Chocolate Chip Recipe -Evolution of Music Streaming -Remote Work Productivity Tips -Artificial Intelligence in Healthcare -🎮 Video Game Development Insights - - -{{MESSAGES:END:2}} -""" - - content = title_generation_template( - template, - form_data["messages"], - { - "name": user.name, - "location": user.info.get("location") if user.info else None, - }, - ) - - payload = { - "model": task_model_id, - "messages": [{"role": "user", "content": content}], - "stream": False, - **( - {"max_tokens": 50} - if models[task_model_id]["owned_by"] == "ollama" - else { - "max_completion_tokens": 50, - } - ), - "metadata": { - "task": str(TASKS.TITLE_GENERATION), - "task_body": form_data, - "chat_id": form_data.get("chat_id", None), - }, - } - - # Handle pipeline filters +async def chat_action( + request: Request, action_id: str, form_data: dict, user=Depends(get_verified_user) +): try: - payload = filter_pipeline(payload, user, models) + return await chat_action_handler(request, action_id, form_data, user) except Exception as e: - if len(e.args) > 1: - return JSONResponse( - status_code=e.args[0], - content={"detail": e.args[1]}, - ) - else: - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - if "chat_id" in payload: - del payload["chat_id"] - - return await generate_chat_completions(form_data=payload, user=user) - - -@app.post("/api/task/tags/completions") -async def generate_chat_tags(form_data: dict, user=Depends(get_verified_user)): - - if not app.state.config.ENABLE_TAGS_GENERATION: - return JSONResponse( - status_code=status.HTTP_200_OK, - content={"detail": "Tags generation is disabled"}, - ) - - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} - - model_id = form_data["model"] - if model_id not in models: - raise HTTPException( - status_code=status.HTTP_404_NOT_FOUND, - detail="Model not found", - ) - - # Check if the user has a custom task model - # If the user has a custom task model, use that model - task_model_id = get_task_model_id( - model_id, - app.state.config.TASK_MODEL, - app.state.config.TASK_MODEL_EXTERNAL, - models, - ) - - log.debug( - f"generating chat tags using model {task_model_id} for user {user.email} " - ) - - if app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE != "": - template = app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE - else: - template = """### Task: -Generate 1-3 broad tags categorizing the main themes of the chat history, along with 1-3 more specific subtopic tags. - -### Guidelines: -- Start with high-level domains (e.g. Science, Technology, Philosophy, Arts, Politics, Business, Health, Sports, Entertainment, Education) -- Consider including relevant subfields/subdomains if they are strongly represented throughout the conversation -- If content is too short (less than 3 messages) or too diverse, use only ["General"] -- Use the chat's primary language; default to English if multilingual -- Prioritize accuracy over specificity - -### Output: -JSON format: { "tags": ["tag1", "tag2", "tag3"] } - -### Chat History: - -{{MESSAGES:END:6}} -""" - - content = tags_generation_template( - template, form_data["messages"], {"name": user.name} - ) - - payload = { - "model": task_model_id, - "messages": [{"role": "user", "content": content}], - "stream": False, - "metadata": { - "task": str(TASKS.TAGS_GENERATION), - "task_body": form_data, - "chat_id": form_data.get("chat_id", None), - }, - } - - # Handle pipeline filters - try: - payload = filter_pipeline(payload, user, models) - except Exception as e: - if len(e.args) > 1: - return JSONResponse( - status_code=e.args[0], - content={"detail": e.args[1]}, - ) - else: - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - if "chat_id" in payload: - del payload["chat_id"] - - return await generate_chat_completions(form_data=payload, user=user) - - -@app.post("/api/task/queries/completions") -async def generate_queries(form_data: dict, user=Depends(get_verified_user)): - - type = form_data.get("type") - if type == "web_search": - if not app.state.config.ENABLE_SEARCH_QUERY_GENERATION: - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=f"Search query generation is disabled", - ) - elif type == "retrieval": - if not app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION: - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=f"Query generation is disabled", - ) - - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} - - model_id = form_data["model"] - if model_id not in models: - raise HTTPException( - status_code=status.HTTP_404_NOT_FOUND, - detail="Model not found", - ) - - # Check if the user has a custom task model - # If the user has a custom task model, use that model - task_model_id = get_task_model_id( - model_id, - app.state.config.TASK_MODEL, - app.state.config.TASK_MODEL_EXTERNAL, - models, - ) - - log.debug( - f"generating {type} queries using model {task_model_id} for user {user.email}" - ) - - if (app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE).strip() != "": - template = app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE - else: - template = DEFAULT_QUERY_GENERATION_PROMPT_TEMPLATE - - content = query_generation_template( - template, form_data["messages"], {"name": user.name} - ) - - payload = { - "model": task_model_id, - "messages": [{"role": "user", "content": content}], - "stream": False, - "metadata": { - "task": str(TASKS.QUERY_GENERATION), - "task_body": form_data, - "chat_id": form_data.get("chat_id", None), - }, - } - - # Handle pipeline filters - try: - payload = filter_pipeline(payload, user, models) - except Exception as e: - if len(e.args) > 1: - return JSONResponse( - status_code=e.args[0], - content={"detail": e.args[1]}, - ) - else: - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - if "chat_id" in payload: - del payload["chat_id"] - - return await generate_chat_completions(form_data=payload, user=user) - - -@app.post("/api/task/auto/completions") -async def generate_autocompletion(form_data: dict, user=Depends(get_verified_user)): - if not app.state.config.ENABLE_AUTOCOMPLETE_GENERATION: raise HTTPException( status_code=status.HTTP_400_BAD_REQUEST, - detail=f"Autocompletion generation is disabled", - ) - - type = form_data.get("type") - prompt = form_data.get("prompt") - messages = form_data.get("messages") - - if app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH > 0: - if len(prompt) > app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH: - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=f"Input prompt exceeds maximum length of {app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH}", - ) - - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} - - model_id = form_data["model"] - if model_id not in models: - raise HTTPException( - status_code=status.HTTP_404_NOT_FOUND, - detail="Model not found", - ) - - # Check if the user has a custom task model - # If the user has a custom task model, use that model - task_model_id = get_task_model_id( - model_id, - app.state.config.TASK_MODEL, - app.state.config.TASK_MODEL_EXTERNAL, - models, - ) - - log.debug( - f"generating autocompletion using model {task_model_id} for user {user.email}" - ) - - if (app.state.config.AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE).strip() != "": - template = app.state.config.AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE - else: - template = DEFAULT_AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE - - content = autocomplete_generation_template( - template, prompt, messages, type, {"name": user.name} - ) - - payload = { - "model": task_model_id, - "messages": [{"role": "user", "content": content}], - "stream": False, - "metadata": { - "task": str(TASKS.AUTOCOMPLETE_GENERATION), - "task_body": form_data, - "chat_id": form_data.get("chat_id", None), - }, - } - - # Handle pipeline filters - try: - payload = filter_pipeline(payload, user, models) - except Exception as e: - if len(e.args) > 1: - return JSONResponse( - status_code=e.args[0], - content={"detail": e.args[1]}, - ) - else: - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - if "chat_id" in payload: - del payload["chat_id"] - - return await generate_chat_completions(form_data=payload, user=user) - - -@app.post("/api/task/emoji/completions") -async def generate_emoji(form_data: dict, user=Depends(get_verified_user)): - - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} - - model_id = form_data["model"] - if model_id not in models: - raise HTTPException( - status_code=status.HTTP_404_NOT_FOUND, - detail="Model not found", - ) - - # Check if the user has a custom task model - # If the user has a custom task model, use that model - task_model_id = get_task_model_id( - model_id, - app.state.config.TASK_MODEL, - app.state.config.TASK_MODEL_EXTERNAL, - models, - ) - - log.debug(f"generating emoji using model {task_model_id} for user {user.email} ") - - template = ''' -Your task is to reflect the speaker's likely facial expression through a fitting emoji. Interpret emotions from the message and reflect their facial expression using fitting, diverse emojis (e.g., 😊, 😢, 😡, 😱). - -Message: """{{prompt}}""" -''' - content = emoji_generation_template( - template, - form_data["prompt"], - { - "name": user.name, - "location": user.info.get("location") if user.info else None, - }, - ) - - payload = { - "model": task_model_id, - "messages": [{"role": "user", "content": content}], - "stream": False, - **( - {"max_tokens": 4} - if models[task_model_id]["owned_by"] == "ollama" - else { - "max_completion_tokens": 4, - } - ), - "chat_id": form_data.get("chat_id", None), - "metadata": {"task": str(TASKS.EMOJI_GENERATION), "task_body": form_data}, - } - - # Handle pipeline filters - try: - payload = filter_pipeline(payload, user, models) - except Exception as e: - if len(e.args) > 1: - return JSONResponse( - status_code=e.args[0], - content={"detail": e.args[1]}, - ) - else: - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - if "chat_id" in payload: - del payload["chat_id"] - - return await generate_chat_completions(form_data=payload, user=user) - - -@app.post("/api/task/moa/completions") -async def generate_moa_response(form_data: dict, user=Depends(get_verified_user)): - - model_list = await get_all_models() - models = {model["id"]: model for model in model_list} - - model_id = form_data["model"] - if model_id not in models: - raise HTTPException( - status_code=status.HTTP_404_NOT_FOUND, - detail="Model not found", - ) - - # Check if the user has a custom task model - # If the user has a custom task model, use that model - task_model_id = get_task_model_id( - model_id, - app.state.config.TASK_MODEL, - app.state.config.TASK_MODEL_EXTERNAL, - models, - ) - - log.debug(f"generating MOA model {task_model_id} for user {user.email} ") - - template = """You have been provided with a set of responses from various models to the latest user query: "{{prompt}}" - -Your task is to synthesize these responses into a single, high-quality response. It is crucial to critically evaluate the information provided in these responses, recognizing that some of it may be biased or incorrect. Your response should not simply replicate the given answers but should offer a refined, accurate, and comprehensive reply to the instruction. Ensure your response is well-structured, coherent, and adheres to the highest standards of accuracy and reliability. - -Responses from models: {{responses}}""" - - content = moa_response_generation_template( - template, - form_data["prompt"], - form_data["responses"], - ) - - payload = { - "model": task_model_id, - "messages": [{"role": "user", "content": content}], - "stream": form_data.get("stream", False), - "chat_id": form_data.get("chat_id", None), - "metadata": { - "task": str(TASKS.MOA_RESPONSE_GENERATION), - "task_body": form_data, - }, - } - - try: - payload = filter_pipeline(payload, user, models) - except Exception as e: - if len(e.args) > 1: - return JSONResponse( - status_code=e.args[0], - content={"detail": e.args[1]}, - ) - else: - return JSONResponse( - status_code=status.HTTP_400_BAD_REQUEST, - content={"detail": str(e)}, - ) - if "chat_id" in payload: - del payload["chat_id"] - - return await generate_chat_completions(form_data=payload, user=user) - - -################################## -# -# Pipelines Endpoints -# -################################## - - -# TODO: Refactor pipelines API endpoints below into a separate file - - -@app.get("/api/pipelines/list") -async def get_pipelines_list(user=Depends(get_admin_user)): - responses = await get_openai_models_responses() - - log.debug(f"get_pipelines_list: get_openai_models_responses returned {responses}") - urlIdxs = [ - idx - for idx, response in enumerate(responses) - if response is not None and "pipelines" in response - ] - - return { - "data": [ - { - "url": openai_app.state.config.OPENAI_API_BASE_URLS[urlIdx], - "idx": urlIdx, - } - for urlIdx in urlIdxs - ] - } - - -@app.post("/api/pipelines/upload") -async def upload_pipeline( - urlIdx: int = Form(...), file: UploadFile = File(...), user=Depends(get_admin_user) -): - print("upload_pipeline", urlIdx, file.filename) - # Check if the uploaded file is a python file - if not (file.filename and file.filename.endswith(".py")): - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail="Only Python (.py) files are allowed.", - ) - - upload_folder = f"{CACHE_DIR}/pipelines" - os.makedirs(upload_folder, exist_ok=True) - file_path = os.path.join(upload_folder, file.filename) - - r = None - try: - # Save the uploaded file - with open(file_path, "wb") as buffer: - shutil.copyfileobj(file.file, buffer) - - url = openai_app.state.config.OPENAI_API_BASE_URLS[urlIdx] - key = openai_app.state.config.OPENAI_API_KEYS[urlIdx] - - headers = {"Authorization": f"Bearer {key}"} - - with open(file_path, "rb") as f: - files = {"file": f} - r = requests.post(f"{url}/pipelines/upload", headers=headers, files=files) - - r.raise_for_status() - data = r.json() - - return {**data} - except Exception as e: - # Handle connection error here - print(f"Connection error: {e}") - - detail = "Pipeline not found" - status_code = status.HTTP_404_NOT_FOUND - if r is not None: - status_code = r.status_code - try: - res = r.json() - if "detail" in res: - detail = res["detail"] - except Exception: - pass - - raise HTTPException( - status_code=status_code, - detail=detail, - ) - finally: - # Ensure the file is deleted after the upload is completed or on failure - if os.path.exists(file_path): - os.remove(file_path) - - -class AddPipelineForm(BaseModel): - url: str - urlIdx: int - - -@app.post("/api/pipelines/add") -async def add_pipeline(form_data: AddPipelineForm, user=Depends(get_admin_user)): - r = None - try: - urlIdx = form_data.urlIdx - - url = openai_app.state.config.OPENAI_API_BASE_URLS[urlIdx] - key = openai_app.state.config.OPENAI_API_KEYS[urlIdx] - - headers = {"Authorization": f"Bearer {key}"} - r = requests.post( - f"{url}/pipelines/add", headers=headers, json={"url": form_data.url} - ) - - r.raise_for_status() - data = r.json() - - return {**data} - except Exception as e: - # Handle connection error here - print(f"Connection error: {e}") - - detail = "Pipeline not found" - if r is not None: - try: - res = r.json() - if "detail" in res: - detail = res["detail"] - except Exception: - pass - - raise HTTPException( - status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), - detail=detail, - ) - - -class DeletePipelineForm(BaseModel): - id: str - urlIdx: int - - -@app.delete("/api/pipelines/delete") -async def delete_pipeline(form_data: DeletePipelineForm, user=Depends(get_admin_user)): - r = None - try: - urlIdx = form_data.urlIdx - - url = openai_app.state.config.OPENAI_API_BASE_URLS[urlIdx] - key = openai_app.state.config.OPENAI_API_KEYS[urlIdx] - - headers = {"Authorization": f"Bearer {key}"} - r = requests.delete( - f"{url}/pipelines/delete", headers=headers, json={"id": form_data.id} - ) - - r.raise_for_status() - data = r.json() - - return {**data} - except Exception as e: - # Handle connection error here - print(f"Connection error: {e}") - - detail = "Pipeline not found" - if r is not None: - try: - res = r.json() - if "detail" in res: - detail = res["detail"] - except Exception: - pass - - raise HTTPException( - status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), - detail=detail, - ) - - -@app.get("/api/pipelines") -async def get_pipelines(urlIdx: Optional[int] = None, user=Depends(get_admin_user)): - r = None - try: - url = openai_app.state.config.OPENAI_API_BASE_URLS[urlIdx] - key = openai_app.state.config.OPENAI_API_KEYS[urlIdx] - - headers = {"Authorization": f"Bearer {key}"} - r = requests.get(f"{url}/pipelines", headers=headers) - - r.raise_for_status() - data = r.json() - - return {**data} - except Exception as e: - # Handle connection error here - print(f"Connection error: {e}") - - detail = "Pipeline not found" - if r is not None: - try: - res = r.json() - if "detail" in res: - detail = res["detail"] - except Exception: - pass - - raise HTTPException( - status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), - detail=detail, - ) - - -@app.get("/api/pipelines/{pipeline_id}/valves") -async def get_pipeline_valves( - urlIdx: Optional[int], - pipeline_id: str, - user=Depends(get_admin_user), -): - r = None - try: - url = openai_app.state.config.OPENAI_API_BASE_URLS[urlIdx] - key = openai_app.state.config.OPENAI_API_KEYS[urlIdx] - - headers = {"Authorization": f"Bearer {key}"} - r = requests.get(f"{url}/{pipeline_id}/valves", headers=headers) - - r.raise_for_status() - data = r.json() - - return {**data} - except Exception as e: - # Handle connection error here - print(f"Connection error: {e}") - - detail = "Pipeline not found" - - if r is not None: - try: - res = r.json() - if "detail" in res: - detail = res["detail"] - except Exception: - pass - - raise HTTPException( - status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), - detail=detail, - ) - - -@app.get("/api/pipelines/{pipeline_id}/valves/spec") -async def get_pipeline_valves_spec( - urlIdx: Optional[int], - pipeline_id: str, - user=Depends(get_admin_user), -): - r = None - try: - url = openai_app.state.config.OPENAI_API_BASE_URLS[urlIdx] - key = openai_app.state.config.OPENAI_API_KEYS[urlIdx] - - headers = {"Authorization": f"Bearer {key}"} - r = requests.get(f"{url}/{pipeline_id}/valves/spec", headers=headers) - - r.raise_for_status() - data = r.json() - - return {**data} - except Exception as e: - # Handle connection error here - print(f"Connection error: {e}") - - detail = "Pipeline not found" - if r is not None: - try: - res = r.json() - if "detail" in res: - detail = res["detail"] - except Exception: - pass - - raise HTTPException( - status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), - detail=detail, - ) - - -@app.post("/api/pipelines/{pipeline_id}/valves/update") -async def update_pipeline_valves( - urlIdx: Optional[int], - pipeline_id: str, - form_data: dict, - user=Depends(get_admin_user), -): - r = None - try: - url = openai_app.state.config.OPENAI_API_BASE_URLS[urlIdx] - key = openai_app.state.config.OPENAI_API_KEYS[urlIdx] - - headers = {"Authorization": f"Bearer {key}"} - r = requests.post( - f"{url}/{pipeline_id}/valves/update", - headers=headers, - json={**form_data}, - ) - - r.raise_for_status() - data = r.json() - - return {**data} - except Exception as e: - # Handle connection error here - print(f"Connection error: {e}") - - detail = "Pipeline not found" - - if r is not None: - try: - res = r.json() - if "detail" in res: - detail = res["detail"] - except Exception: - pass - - raise HTTPException( - status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), - detail=detail, + detail=str(e), ) @@ -2602,17 +920,17 @@ async def get_app_config(request: Request): }, "features": { "auth": WEBUI_AUTH, - "auth_trusted_header": bool(webui_app.state.AUTH_TRUSTED_EMAIL_HEADER), - "enable_ldap": webui_app.state.config.ENABLE_LDAP, - "enable_api_key": webui_app.state.config.ENABLE_API_KEY, - "enable_signup": webui_app.state.config.ENABLE_SIGNUP, - "enable_login_form": webui_app.state.config.ENABLE_LOGIN_FORM, + "auth_trusted_header": bool(app.state.AUTH_TRUSTED_EMAIL_HEADER), + "enable_ldap": app.state.config.ENABLE_LDAP, + "enable_api_key": app.state.config.ENABLE_API_KEY, + "enable_signup": app.state.config.ENABLE_SIGNUP, + "enable_login_form": app.state.config.ENABLE_LOGIN_FORM, **( { - "enable_web_search": retrieval_app.state.config.ENABLE_RAG_WEB_SEARCH, - "enable_image_generation": images_app.state.config.ENABLED, - "enable_community_sharing": webui_app.state.config.ENABLE_COMMUNITY_SHARING, - "enable_message_rating": webui_app.state.config.ENABLE_MESSAGE_RATING, + "enable_web_search": app.state.config.ENABLE_RAG_WEB_SEARCH, + "enable_image_generation": app.state.config.ENABLE_IMAGE_GENERATION, + "enable_community_sharing": app.state.config.ENABLE_COMMUNITY_SHARING, + "enable_message_rating": app.state.config.ENABLE_MESSAGE_RATING, "enable_admin_export": ENABLE_ADMIN_EXPORT, "enable_admin_chat_access": ENABLE_ADMIN_CHAT_ACCESS, } @@ -2622,23 +940,23 @@ async def get_app_config(request: Request): }, **( { - "default_models": webui_app.state.config.DEFAULT_MODELS, - "default_prompt_suggestions": webui_app.state.config.DEFAULT_PROMPT_SUGGESTIONS, + "default_models": app.state.config.DEFAULT_MODELS, + "default_prompt_suggestions": app.state.config.DEFAULT_PROMPT_SUGGESTIONS, "audio": { "tts": { - "engine": audio_app.state.config.TTS_ENGINE, - "voice": audio_app.state.config.TTS_VOICE, - "split_on": audio_app.state.config.TTS_SPLIT_ON, + "engine": app.state.config.TTS_ENGINE, + "voice": app.state.config.TTS_VOICE, + "split_on": app.state.config.TTS_SPLIT_ON, }, "stt": { - "engine": audio_app.state.config.STT_ENGINE, + "engine": app.state.config.STT_ENGINE, }, }, "file": { - "max_size": retrieval_app.state.config.FILE_MAX_SIZE, - "max_count": retrieval_app.state.config.FILE_MAX_COUNT, + "max_size": app.state.config.FILE_MAX_SIZE, + "max_count": app.state.config.FILE_MAX_COUNT, }, - "permissions": {**webui_app.state.config.USER_PERMISSIONS}, + "permissions": {**app.state.config.USER_PERMISSIONS}, } if user is not None else {} @@ -2646,7 +964,8 @@ async def get_app_config(request: Request): } -# TODO: webhook endpoint should be under config endpoints +class UrlForm(BaseModel): + url: str @app.get("/api/webhook") @@ -2656,14 +975,10 @@ async def get_webhook_url(user=Depends(get_admin_user)): } -class UrlForm(BaseModel): - url: str - - @app.post("/api/webhook") async def update_webhook_url(form_data: UrlForm, user=Depends(get_admin_user)): app.state.config.WEBHOOK_URL = form_data.url - webui_app.state.WEBHOOK_URL = app.state.config.WEBHOOK_URL + app.state.WEBHOOK_URL = app.state.config.WEBHOOK_URL return {"url": app.state.config.WEBHOOK_URL} @@ -2674,11 +989,6 @@ async def get_app_version(): } -@app.get("/api/changelog") -async def get_app_changelog(): - return {key: CHANGELOG[key] for idx, key in enumerate(CHANGELOG) if idx < 5} - - @app.get("/api/version/updates") async def get_app_latest_release_version(): if OFFLINE_MODE: @@ -2702,6 +1012,11 @@ async def get_app_latest_release_version(): return {"current": VERSION, "latest": VERSION} +@app.get("/api/changelog") +async def get_app_changelog(): + return {key: CHANGELOG[key] for idx, key in enumerate(CHANGELOG) if idx < 5} + + ############################ # OAuth Login & Callback ############################ @@ -2789,7 +1104,6 @@ async def healthcheck_with_db(): app.mount("/static", StaticFiles(directory=STATIC_DIR), name="static") app.mount("/cache", StaticFiles(directory=CACHE_DIR), name="cache") - if os.path.exists(FRONTEND_BUILD_DIR): mimetypes.add_type("text/javascript", ".js") app.mount( diff --git a/backend/open_webui/migrations/env.py b/backend/open_webui/migrations/env.py index 5e860c8a0..128881647 100644 --- a/backend/open_webui/migrations/env.py +++ b/backend/open_webui/migrations/env.py @@ -1,7 +1,7 @@ from logging.config import fileConfig from alembic import context -from open_webui.apps.webui.models.auths import Auth +from open_webui.models.auths import Auth from open_webui.env import DATABASE_URL from sqlalchemy import engine_from_config, pool diff --git a/backend/open_webui/migrations/script.py.mako b/backend/open_webui/migrations/script.py.mako index 01e730e77..bcf5567fd 100644 --- a/backend/open_webui/migrations/script.py.mako +++ b/backend/open_webui/migrations/script.py.mako @@ -9,7 +9,7 @@ from typing import Sequence, Union from alembic import op import sqlalchemy as sa -import open_webui.apps.webui.internal.db +import open_webui.internal.db ${imports if imports else ""} # revision identifiers, used by Alembic. diff --git a/backend/open_webui/migrations/versions/7e5b5dc7342b_init.py b/backend/open_webui/migrations/versions/7e5b5dc7342b_init.py index 607a7b2c9..9e56282ef 100644 --- a/backend/open_webui/migrations/versions/7e5b5dc7342b_init.py +++ b/backend/open_webui/migrations/versions/7e5b5dc7342b_init.py @@ -11,8 +11,8 @@ from typing import Sequence, Union import sqlalchemy as sa from alembic import op -import open_webui.apps.webui.internal.db -from open_webui.apps.webui.internal.db import JSONField +import open_webui.internal.db +from open_webui.internal.db import JSONField from open_webui.migrations.util import get_existing_tables # revision identifiers, used by Alembic. diff --git a/backend/open_webui/apps/webui/models/auths.py b/backend/open_webui/models/auths.py similarity index 97% rename from backend/open_webui/apps/webui/models/auths.py rename to backend/open_webui/models/auths.py index 391b2e9ec..f07c36c73 100644 --- a/backend/open_webui/apps/webui/models/auths.py +++ b/backend/open_webui/models/auths.py @@ -2,8 +2,8 @@ import logging import uuid from typing import Optional -from open_webui.apps.webui.internal.db import Base, get_db -from open_webui.apps.webui.models.users import UserModel, Users +from open_webui.internal.db import Base, get_db +from open_webui.models.users import UserModel, Users from open_webui.env import SRC_LOG_LEVELS from pydantic import BaseModel from sqlalchemy import Boolean, Column, String, Text diff --git a/backend/open_webui/apps/webui/models/chats.py b/backend/open_webui/models/chats.py similarity index 99% rename from backend/open_webui/apps/webui/models/chats.py rename to backend/open_webui/models/chats.py index 21250add8..3e621a150 100644 --- a/backend/open_webui/apps/webui/models/chats.py +++ b/backend/open_webui/models/chats.py @@ -3,8 +3,8 @@ import time import uuid from typing import Optional -from open_webui.apps.webui.internal.db import Base, get_db -from open_webui.apps.webui.models.tags import TagModel, Tag, Tags +from open_webui.internal.db import Base, get_db +from open_webui.models.tags import TagModel, Tag, Tags from pydantic import BaseModel, ConfigDict diff --git a/backend/open_webui/apps/webui/models/feedbacks.py b/backend/open_webui/models/feedbacks.py similarity index 98% rename from backend/open_webui/apps/webui/models/feedbacks.py rename to backend/open_webui/models/feedbacks.py index c2356dfd8..7ff5c4540 100644 --- a/backend/open_webui/apps/webui/models/feedbacks.py +++ b/backend/open_webui/models/feedbacks.py @@ -3,8 +3,8 @@ import time import uuid from typing import Optional -from open_webui.apps.webui.internal.db import Base, get_db -from open_webui.apps.webui.models.chats import Chats +from open_webui.internal.db import Base, get_db +from open_webui.models.chats import Chats from open_webui.env import SRC_LOG_LEVELS from pydantic import BaseModel, ConfigDict diff --git a/backend/open_webui/apps/webui/models/files.py b/backend/open_webui/models/files.py similarity index 98% rename from backend/open_webui/apps/webui/models/files.py rename to backend/open_webui/models/files.py index 31c9164b6..4050b0140 100644 --- a/backend/open_webui/apps/webui/models/files.py +++ b/backend/open_webui/models/files.py @@ -2,7 +2,7 @@ import logging import time from typing import Optional -from open_webui.apps.webui.internal.db import Base, JSONField, get_db +from open_webui.internal.db import Base, JSONField, get_db from open_webui.env import SRC_LOG_LEVELS from pydantic import BaseModel, ConfigDict from sqlalchemy import BigInteger, Column, String, Text, JSON diff --git a/backend/open_webui/apps/webui/models/folders.py b/backend/open_webui/models/folders.py similarity index 98% rename from backend/open_webui/apps/webui/models/folders.py rename to backend/open_webui/models/folders.py index 90e8880aa..040774196 100644 --- a/backend/open_webui/apps/webui/models/folders.py +++ b/backend/open_webui/models/folders.py @@ -3,8 +3,8 @@ import time import uuid from typing import Optional -from open_webui.apps.webui.internal.db import Base, get_db -from open_webui.apps.webui.models.chats import Chats +from open_webui.internal.db import Base, get_db +from open_webui.models.chats import Chats from open_webui.env import SRC_LOG_LEVELS from pydantic import BaseModel, ConfigDict diff --git a/backend/open_webui/apps/webui/models/functions.py b/backend/open_webui/models/functions.py similarity index 98% rename from backend/open_webui/apps/webui/models/functions.py rename to backend/open_webui/models/functions.py index fda155075..6c6aed862 100644 --- a/backend/open_webui/apps/webui/models/functions.py +++ b/backend/open_webui/models/functions.py @@ -2,8 +2,8 @@ import logging import time from typing import Optional -from open_webui.apps.webui.internal.db import Base, JSONField, get_db -from open_webui.apps.webui.models.users import Users +from open_webui.internal.db import Base, JSONField, get_db +from open_webui.models.users import Users from open_webui.env import SRC_LOG_LEVELS from pydantic import BaseModel, ConfigDict from sqlalchemy import BigInteger, Boolean, Column, String, Text diff --git a/backend/open_webui/apps/webui/models/groups.py b/backend/open_webui/models/groups.py similarity index 97% rename from backend/open_webui/apps/webui/models/groups.py rename to backend/open_webui/models/groups.py index e692198cd..8f0728411 100644 --- a/backend/open_webui/apps/webui/models/groups.py +++ b/backend/open_webui/models/groups.py @@ -4,10 +4,10 @@ import time from typing import Optional import uuid -from open_webui.apps.webui.internal.db import Base, get_db +from open_webui.internal.db import Base, get_db from open_webui.env import SRC_LOG_LEVELS -from open_webui.apps.webui.models.files import FileMetadataResponse +from open_webui.models.files import FileMetadataResponse from pydantic import BaseModel, ConfigDict diff --git a/backend/open_webui/apps/webui/models/knowledge.py b/backend/open_webui/models/knowledge.py similarity index 97% rename from backend/open_webui/apps/webui/models/knowledge.py rename to backend/open_webui/models/knowledge.py index e1a13b3fd..bed3d5542 100644 --- a/backend/open_webui/apps/webui/models/knowledge.py +++ b/backend/open_webui/models/knowledge.py @@ -4,11 +4,11 @@ import time from typing import Optional import uuid -from open_webui.apps.webui.internal.db import Base, get_db +from open_webui.internal.db import Base, get_db from open_webui.env import SRC_LOG_LEVELS -from open_webui.apps.webui.models.files import FileMetadataResponse -from open_webui.apps.webui.models.users import Users, UserResponse +from open_webui.models.files import FileMetadataResponse +from open_webui.models.users import Users, UserResponse from pydantic import BaseModel, ConfigDict diff --git a/backend/open_webui/apps/webui/models/memories.py b/backend/open_webui/models/memories.py similarity index 98% rename from backend/open_webui/apps/webui/models/memories.py rename to backend/open_webui/models/memories.py index 6686058d3..c8dae9726 100644 --- a/backend/open_webui/apps/webui/models/memories.py +++ b/backend/open_webui/models/memories.py @@ -2,7 +2,7 @@ import time import uuid from typing import Optional -from open_webui.apps.webui.internal.db import Base, get_db +from open_webui.internal.db import Base, get_db from pydantic import BaseModel, ConfigDict from sqlalchemy import BigInteger, Column, String, Text diff --git a/backend/open_webui/apps/webui/models/models.py b/backend/open_webui/models/models.py similarity index 98% rename from backend/open_webui/apps/webui/models/models.py rename to backend/open_webui/models/models.py index 50581bc73..f2f59d7c4 100644 --- a/backend/open_webui/apps/webui/models/models.py +++ b/backend/open_webui/models/models.py @@ -2,10 +2,10 @@ import logging import time from typing import Optional -from open_webui.apps.webui.internal.db import Base, JSONField, get_db +from open_webui.internal.db import Base, JSONField, get_db from open_webui.env import SRC_LOG_LEVELS -from open_webui.apps.webui.models.users import Users, UserResponse +from open_webui.models.users import Users, UserResponse from pydantic import BaseModel, ConfigDict diff --git a/backend/open_webui/apps/webui/models/prompts.py b/backend/open_webui/models/prompts.py similarity index 97% rename from backend/open_webui/apps/webui/models/prompts.py rename to backend/open_webui/models/prompts.py index fe9999195..8ef4cd2be 100644 --- a/backend/open_webui/apps/webui/models/prompts.py +++ b/backend/open_webui/models/prompts.py @@ -1,8 +1,8 @@ import time from typing import Optional -from open_webui.apps.webui.internal.db import Base, get_db -from open_webui.apps.webui.models.users import Users, UserResponse +from open_webui.internal.db import Base, get_db +from open_webui.models.users import Users, UserResponse from pydantic import BaseModel, ConfigDict from sqlalchemy import BigInteger, Column, String, Text, JSON diff --git a/backend/open_webui/apps/webui/models/tags.py b/backend/open_webui/models/tags.py similarity index 98% rename from backend/open_webui/apps/webui/models/tags.py rename to backend/open_webui/models/tags.py index 7424a2660..3e812db95 100644 --- a/backend/open_webui/apps/webui/models/tags.py +++ b/backend/open_webui/models/tags.py @@ -3,7 +3,7 @@ import time import uuid from typing import Optional -from open_webui.apps.webui.internal.db import Base, get_db +from open_webui.internal.db import Base, get_db from open_webui.env import SRC_LOG_LEVELS diff --git a/backend/open_webui/apps/webui/models/tools.py b/backend/open_webui/models/tools.py similarity index 98% rename from backend/open_webui/apps/webui/models/tools.py rename to backend/open_webui/models/tools.py index 8f798c317..a5f13ebb7 100644 --- a/backend/open_webui/apps/webui/models/tools.py +++ b/backend/open_webui/models/tools.py @@ -2,8 +2,8 @@ import logging import time from typing import Optional -from open_webui.apps.webui.internal.db import Base, JSONField, get_db -from open_webui.apps.webui.models.users import Users, UserResponse +from open_webui.internal.db import Base, JSONField, get_db +from open_webui.models.users import Users, UserResponse from open_webui.env import SRC_LOG_LEVELS from pydantic import BaseModel, ConfigDict from sqlalchemy import BigInteger, Column, String, Text, JSON diff --git a/backend/open_webui/apps/webui/models/users.py b/backend/open_webui/models/users.py similarity index 98% rename from backend/open_webui/apps/webui/models/users.py rename to backend/open_webui/models/users.py index 5bbcc3099..5b6c27214 100644 --- a/backend/open_webui/apps/webui/models/users.py +++ b/backend/open_webui/models/users.py @@ -1,8 +1,8 @@ import time from typing import Optional -from open_webui.apps.webui.internal.db import Base, JSONField, get_db -from open_webui.apps.webui.models.chats import Chats +from open_webui.internal.db import Base, JSONField, get_db +from open_webui.models.chats import Chats from pydantic import BaseModel, ConfigDict from sqlalchemy import BigInteger, Column, String, Text diff --git a/backend/open_webui/apps/retrieval/loaders/main.py b/backend/open_webui/retrieval/loaders/main.py similarity index 100% rename from backend/open_webui/apps/retrieval/loaders/main.py rename to backend/open_webui/retrieval/loaders/main.py diff --git a/backend/open_webui/apps/retrieval/loaders/youtube.py b/backend/open_webui/retrieval/loaders/youtube.py similarity index 100% rename from backend/open_webui/apps/retrieval/loaders/youtube.py rename to backend/open_webui/retrieval/loaders/youtube.py diff --git a/backend/open_webui/apps/retrieval/models/colbert.py b/backend/open_webui/retrieval/models/colbert.py similarity index 100% rename from backend/open_webui/apps/retrieval/models/colbert.py rename to backend/open_webui/retrieval/models/colbert.py diff --git a/backend/open_webui/apps/retrieval/utils.py b/backend/open_webui/retrieval/utils.py similarity index 99% rename from backend/open_webui/apps/retrieval/utils.py rename to backend/open_webui/retrieval/utils.py index bf939ecf1..9444ade95 100644 --- a/backend/open_webui/apps/retrieval/utils.py +++ b/backend/open_webui/retrieval/utils.py @@ -11,7 +11,7 @@ from langchain.retrievers import ContextualCompressionRetriever, EnsembleRetriev from langchain_community.retrievers import BM25Retriever from langchain_core.documents import Document -from open_webui.apps.retrieval.vector.connector import VECTOR_DB_CLIENT +from open_webui.retrieval.vector.connector import VECTOR_DB_CLIENT from open_webui.utils.misc import get_last_user_message from open_webui.env import SRC_LOG_LEVELS diff --git a/backend/open_webui/retrieval/vector/connector.py b/backend/open_webui/retrieval/vector/connector.py new file mode 100644 index 000000000..bf97bc7b1 --- /dev/null +++ b/backend/open_webui/retrieval/vector/connector.py @@ -0,0 +1,22 @@ +from open_webui.config import VECTOR_DB + +if VECTOR_DB == "milvus": + from open_webui.retrieval.vector.dbs.milvus import MilvusClient + + VECTOR_DB_CLIENT = MilvusClient() +elif VECTOR_DB == "qdrant": + from open_webui.retrieval.vector.dbs.qdrant import QdrantClient + + VECTOR_DB_CLIENT = QdrantClient() +elif VECTOR_DB == "opensearch": + from open_webui.retrieval.vector.dbs.opensearch import OpenSearchClient + + VECTOR_DB_CLIENT = OpenSearchClient() +elif VECTOR_DB == "pgvector": + from open_webui.retrieval.vector.dbs.pgvector import PgvectorClient + + VECTOR_DB_CLIENT = PgvectorClient() +else: + from open_webui.retrieval.vector.dbs.chroma import ChromaClient + + VECTOR_DB_CLIENT = ChromaClient() diff --git a/backend/open_webui/apps/retrieval/vector/dbs/chroma.py b/backend/open_webui/retrieval/vector/dbs/chroma.py similarity index 98% rename from backend/open_webui/apps/retrieval/vector/dbs/chroma.py rename to backend/open_webui/retrieval/vector/dbs/chroma.py index b2fcdd16a..00d73a889 100644 --- a/backend/open_webui/apps/retrieval/vector/dbs/chroma.py +++ b/backend/open_webui/retrieval/vector/dbs/chroma.py @@ -4,7 +4,7 @@ from chromadb.utils.batch_utils import create_batches from typing import Optional -from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult +from open_webui.retrieval.vector.main import VectorItem, SearchResult, GetResult from open_webui.config import ( CHROMA_DATA_PATH, CHROMA_HTTP_HOST, diff --git a/backend/open_webui/apps/retrieval/vector/dbs/milvus.py b/backend/open_webui/retrieval/vector/dbs/milvus.py similarity index 99% rename from backend/open_webui/apps/retrieval/vector/dbs/milvus.py rename to backend/open_webui/retrieval/vector/dbs/milvus.py index 5351f860e..31d890664 100644 --- a/backend/open_webui/apps/retrieval/vector/dbs/milvus.py +++ b/backend/open_webui/retrieval/vector/dbs/milvus.py @@ -4,7 +4,7 @@ import json from typing import Optional -from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult +from open_webui.retrieval.vector.main import VectorItem, SearchResult, GetResult from open_webui.config import ( MILVUS_URI, ) diff --git a/backend/open_webui/apps/retrieval/vector/dbs/opensearch.py b/backend/open_webui/retrieval/vector/dbs/opensearch.py similarity index 98% rename from backend/open_webui/apps/retrieval/vector/dbs/opensearch.py rename to backend/open_webui/retrieval/vector/dbs/opensearch.py index 6234b2837..b3d8b5eb8 100644 --- a/backend/open_webui/apps/retrieval/vector/dbs/opensearch.py +++ b/backend/open_webui/retrieval/vector/dbs/opensearch.py @@ -1,7 +1,7 @@ from opensearchpy import OpenSearch from typing import Optional -from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult +from open_webui.retrieval.vector.main import VectorItem, SearchResult, GetResult from open_webui.config import ( OPENSEARCH_URI, OPENSEARCH_SSL, diff --git a/backend/open_webui/apps/retrieval/vector/dbs/pgvector.py b/backend/open_webui/retrieval/vector/dbs/pgvector.py similarity index 98% rename from backend/open_webui/apps/retrieval/vector/dbs/pgvector.py rename to backend/open_webui/retrieval/vector/dbs/pgvector.py index d537943a1..cb8c545e9 100644 --- a/backend/open_webui/apps/retrieval/vector/dbs/pgvector.py +++ b/backend/open_webui/retrieval/vector/dbs/pgvector.py @@ -18,7 +18,7 @@ from sqlalchemy.dialects.postgresql import JSONB, array from pgvector.sqlalchemy import Vector from sqlalchemy.ext.mutable import MutableDict -from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult +from open_webui.retrieval.vector.main import VectorItem, SearchResult, GetResult from open_webui.config import PGVECTOR_DB_URL VECTOR_LENGTH = 1536 @@ -40,7 +40,7 @@ class PgvectorClient: # if no pgvector uri, use the existing database connection if not PGVECTOR_DB_URL: - from open_webui.apps.webui.internal.db import Session + from open_webui.internal.db import Session self.session = Session else: diff --git a/backend/open_webui/apps/retrieval/vector/dbs/qdrant.py b/backend/open_webui/retrieval/vector/dbs/qdrant.py similarity index 98% rename from backend/open_webui/apps/retrieval/vector/dbs/qdrant.py rename to backend/open_webui/retrieval/vector/dbs/qdrant.py index 60c1c3d4d..f077ae45a 100644 --- a/backend/open_webui/apps/retrieval/vector/dbs/qdrant.py +++ b/backend/open_webui/retrieval/vector/dbs/qdrant.py @@ -4,7 +4,7 @@ from qdrant_client import QdrantClient as Qclient from qdrant_client.http.models import PointStruct from qdrant_client.models import models -from open_webui.apps.retrieval.vector.main import VectorItem, SearchResult, GetResult +from open_webui.retrieval.vector.main import VectorItem, SearchResult, GetResult from open_webui.config import QDRANT_URI, QDRANT_API_KEY NO_LIMIT = 999999999 diff --git a/backend/open_webui/apps/retrieval/vector/main.py b/backend/open_webui/retrieval/vector/main.py similarity index 100% rename from backend/open_webui/apps/retrieval/vector/main.py rename to backend/open_webui/retrieval/vector/main.py diff --git a/backend/open_webui/apps/retrieval/web/bing.py b/backend/open_webui/retrieval/web/bing.py similarity index 96% rename from backend/open_webui/apps/retrieval/web/bing.py rename to backend/open_webui/retrieval/web/bing.py index b5f889c54..09beb3460 100644 --- a/backend/open_webui/apps/retrieval/web/bing.py +++ b/backend/open_webui/retrieval/web/bing.py @@ -3,7 +3,7 @@ import os from pprint import pprint from typing import Optional import requests -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from open_webui.env import SRC_LOG_LEVELS import argparse diff --git a/backend/open_webui/apps/retrieval/web/brave.py b/backend/open_webui/retrieval/web/brave.py similarity index 93% rename from backend/open_webui/apps/retrieval/web/brave.py rename to backend/open_webui/retrieval/web/brave.py index f988b3b08..3075db990 100644 --- a/backend/open_webui/apps/retrieval/web/brave.py +++ b/backend/open_webui/retrieval/web/brave.py @@ -2,7 +2,7 @@ import logging from typing import Optional import requests -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from open_webui.env import SRC_LOG_LEVELS log = logging.getLogger(__name__) diff --git a/backend/open_webui/apps/retrieval/web/duckduckgo.py b/backend/open_webui/retrieval/web/duckduckgo.py similarity index 95% rename from backend/open_webui/apps/retrieval/web/duckduckgo.py rename to backend/open_webui/retrieval/web/duckduckgo.py index 11e512296..7c0c3f1c2 100644 --- a/backend/open_webui/apps/retrieval/web/duckduckgo.py +++ b/backend/open_webui/retrieval/web/duckduckgo.py @@ -1,7 +1,7 @@ import logging from typing import Optional -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from duckduckgo_search import DDGS from open_webui.env import SRC_LOG_LEVELS diff --git a/backend/open_webui/apps/retrieval/web/google_pse.py b/backend/open_webui/retrieval/web/google_pse.py similarity index 94% rename from backend/open_webui/apps/retrieval/web/google_pse.py rename to backend/open_webui/retrieval/web/google_pse.py index 61b919583..2c51dd3c9 100644 --- a/backend/open_webui/apps/retrieval/web/google_pse.py +++ b/backend/open_webui/retrieval/web/google_pse.py @@ -2,7 +2,7 @@ import logging from typing import Optional import requests -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from open_webui.env import SRC_LOG_LEVELS log = logging.getLogger(__name__) diff --git a/backend/open_webui/apps/retrieval/web/jina_search.py b/backend/open_webui/retrieval/web/jina_search.py similarity index 94% rename from backend/open_webui/apps/retrieval/web/jina_search.py rename to backend/open_webui/retrieval/web/jina_search.py index f5e2febbe..3de6c1807 100644 --- a/backend/open_webui/apps/retrieval/web/jina_search.py +++ b/backend/open_webui/retrieval/web/jina_search.py @@ -1,7 +1,7 @@ import logging import requests -from open_webui.apps.retrieval.web.main import SearchResult +from open_webui.retrieval.web.main import SearchResult from open_webui.env import SRC_LOG_LEVELS from yarl import URL diff --git a/backend/open_webui/apps/retrieval/web/kagi.py b/backend/open_webui/retrieval/web/kagi.py similarity index 86% rename from backend/open_webui/apps/retrieval/web/kagi.py rename to backend/open_webui/retrieval/web/kagi.py index c8c2699ed..0b69da8bc 100644 --- a/backend/open_webui/apps/retrieval/web/kagi.py +++ b/backend/open_webui/retrieval/web/kagi.py @@ -2,7 +2,7 @@ import logging from typing import Optional import requests -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from open_webui.env import SRC_LOG_LEVELS log = logging.getLogger(__name__) @@ -31,17 +31,15 @@ def search_kagi( response.raise_for_status() json_response = response.json() search_results = json_response.get("data", []) - + results = [ SearchResult( - link=result["url"], - title=result["title"], - snippet=result.get("snippet") + link=result["url"], title=result["title"], snippet=result.get("snippet") ) for result in search_results if result["t"] == 0 ] - + print(results) if filter_list: diff --git a/backend/open_webui/apps/retrieval/web/main.py b/backend/open_webui/retrieval/web/main.py similarity index 100% rename from backend/open_webui/apps/retrieval/web/main.py rename to backend/open_webui/retrieval/web/main.py diff --git a/backend/open_webui/apps/retrieval/web/mojeek.py b/backend/open_webui/retrieval/web/mojeek.py similarity index 93% rename from backend/open_webui/apps/retrieval/web/mojeek.py rename to backend/open_webui/retrieval/web/mojeek.py index f257c92aa..d298b0ee5 100644 --- a/backend/open_webui/apps/retrieval/web/mojeek.py +++ b/backend/open_webui/retrieval/web/mojeek.py @@ -2,7 +2,7 @@ import logging from typing import Optional import requests -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from open_webui.env import SRC_LOG_LEVELS log = logging.getLogger(__name__) diff --git a/backend/open_webui/apps/retrieval/web/searchapi.py b/backend/open_webui/retrieval/web/searchapi.py similarity index 93% rename from backend/open_webui/apps/retrieval/web/searchapi.py rename to backend/open_webui/retrieval/web/searchapi.py index 412dc6b69..38bc0b574 100644 --- a/backend/open_webui/apps/retrieval/web/searchapi.py +++ b/backend/open_webui/retrieval/web/searchapi.py @@ -3,7 +3,7 @@ from typing import Optional from urllib.parse import urlencode import requests -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from open_webui.env import SRC_LOG_LEVELS log = logging.getLogger(__name__) diff --git a/backend/open_webui/apps/retrieval/web/searxng.py b/backend/open_webui/retrieval/web/searxng.py similarity index 97% rename from backend/open_webui/apps/retrieval/web/searxng.py rename to backend/open_webui/retrieval/web/searxng.py index cb1eaf91d..15e3c098a 100644 --- a/backend/open_webui/apps/retrieval/web/searxng.py +++ b/backend/open_webui/retrieval/web/searxng.py @@ -2,7 +2,7 @@ import logging from typing import Optional import requests -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from open_webui.env import SRC_LOG_LEVELS log = logging.getLogger(__name__) diff --git a/backend/open_webui/apps/retrieval/web/serper.py b/backend/open_webui/retrieval/web/serper.py similarity index 93% rename from backend/open_webui/apps/retrieval/web/serper.py rename to backend/open_webui/retrieval/web/serper.py index 436fa167e..685e34375 100644 --- a/backend/open_webui/apps/retrieval/web/serper.py +++ b/backend/open_webui/retrieval/web/serper.py @@ -3,7 +3,7 @@ import logging from typing import Optional import requests -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from open_webui.env import SRC_LOG_LEVELS log = logging.getLogger(__name__) diff --git a/backend/open_webui/apps/retrieval/web/serply.py b/backend/open_webui/retrieval/web/serply.py similarity index 95% rename from backend/open_webui/apps/retrieval/web/serply.py rename to backend/open_webui/retrieval/web/serply.py index 1c2521c47..a9b473eb0 100644 --- a/backend/open_webui/apps/retrieval/web/serply.py +++ b/backend/open_webui/retrieval/web/serply.py @@ -3,7 +3,7 @@ from typing import Optional from urllib.parse import urlencode import requests -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from open_webui.env import SRC_LOG_LEVELS log = logging.getLogger(__name__) diff --git a/backend/open_webui/apps/retrieval/web/serpstack.py b/backend/open_webui/retrieval/web/serpstack.py similarity index 94% rename from backend/open_webui/apps/retrieval/web/serpstack.py rename to backend/open_webui/retrieval/web/serpstack.py index b655934de..d4dbda57c 100644 --- a/backend/open_webui/apps/retrieval/web/serpstack.py +++ b/backend/open_webui/retrieval/web/serpstack.py @@ -2,7 +2,7 @@ import logging from typing import Optional import requests -from open_webui.apps.retrieval.web.main import SearchResult, get_filtered_results +from open_webui.retrieval.web.main import SearchResult, get_filtered_results from open_webui.env import SRC_LOG_LEVELS log = logging.getLogger(__name__) diff --git a/backend/open_webui/apps/retrieval/web/tavily.py b/backend/open_webui/retrieval/web/tavily.py similarity index 94% rename from backend/open_webui/apps/retrieval/web/tavily.py rename to backend/open_webui/retrieval/web/tavily.py index 03b0be75a..cc468725d 100644 --- a/backend/open_webui/apps/retrieval/web/tavily.py +++ b/backend/open_webui/retrieval/web/tavily.py @@ -1,7 +1,7 @@ import logging import requests -from open_webui.apps.retrieval.web.main import SearchResult +from open_webui.retrieval.web.main import SearchResult from open_webui.env import SRC_LOG_LEVELS log = logging.getLogger(__name__) diff --git a/backend/open_webui/apps/retrieval/web/testdata/bing.json b/backend/open_webui/retrieval/web/testdata/bing.json similarity index 100% rename from backend/open_webui/apps/retrieval/web/testdata/bing.json rename to backend/open_webui/retrieval/web/testdata/bing.json diff --git a/backend/open_webui/apps/retrieval/web/testdata/brave.json b/backend/open_webui/retrieval/web/testdata/brave.json similarity index 100% rename from backend/open_webui/apps/retrieval/web/testdata/brave.json rename to backend/open_webui/retrieval/web/testdata/brave.json diff --git a/backend/open_webui/apps/retrieval/web/testdata/google_pse.json b/backend/open_webui/retrieval/web/testdata/google_pse.json similarity index 100% rename from backend/open_webui/apps/retrieval/web/testdata/google_pse.json rename to backend/open_webui/retrieval/web/testdata/google_pse.json diff --git a/backend/open_webui/apps/retrieval/web/testdata/searchapi.json b/backend/open_webui/retrieval/web/testdata/searchapi.json similarity index 100% rename from backend/open_webui/apps/retrieval/web/testdata/searchapi.json rename to backend/open_webui/retrieval/web/testdata/searchapi.json diff --git a/backend/open_webui/apps/retrieval/web/testdata/searxng.json b/backend/open_webui/retrieval/web/testdata/searxng.json similarity index 100% rename from backend/open_webui/apps/retrieval/web/testdata/searxng.json rename to backend/open_webui/retrieval/web/testdata/searxng.json diff --git a/backend/open_webui/apps/retrieval/web/testdata/serper.json b/backend/open_webui/retrieval/web/testdata/serper.json similarity index 100% rename from backend/open_webui/apps/retrieval/web/testdata/serper.json rename to backend/open_webui/retrieval/web/testdata/serper.json diff --git a/backend/open_webui/apps/retrieval/web/testdata/serply.json b/backend/open_webui/retrieval/web/testdata/serply.json similarity index 100% rename from backend/open_webui/apps/retrieval/web/testdata/serply.json rename to backend/open_webui/retrieval/web/testdata/serply.json diff --git a/backend/open_webui/apps/retrieval/web/testdata/serpstack.json b/backend/open_webui/retrieval/web/testdata/serpstack.json similarity index 100% rename from backend/open_webui/apps/retrieval/web/testdata/serpstack.json rename to backend/open_webui/retrieval/web/testdata/serpstack.json diff --git a/backend/open_webui/apps/retrieval/web/utils.py b/backend/open_webui/retrieval/web/utils.py similarity index 100% rename from backend/open_webui/apps/retrieval/web/utils.py rename to backend/open_webui/retrieval/web/utils.py diff --git a/backend/open_webui/routers/audio.py b/backend/open_webui/routers/audio.py new file mode 100644 index 000000000..a26355945 --- /dev/null +++ b/backend/open_webui/routers/audio.py @@ -0,0 +1,703 @@ +import hashlib +import json +import logging +import os +import uuid +from functools import lru_cache +from pathlib import Path +from pydub import AudioSegment +from pydub.silence import split_on_silence + +import aiohttp +import aiofiles +import requests + +from fastapi import ( + Depends, + FastAPI, + File, + HTTPException, + Request, + UploadFile, + status, + APIRouter, +) +from fastapi.middleware.cors import CORSMiddleware +from fastapi.responses import FileResponse +from pydantic import BaseModel + + +from open_webui.utils.auth import get_admin_user, get_verified_user +from open_webui.config import ( + WHISPER_MODEL_AUTO_UPDATE, + WHISPER_MODEL_DIR, + CACHE_DIR, +) + +from open_webui.constants import ERROR_MESSAGES +from open_webui.env import ( + ENV, + SRC_LOG_LEVELS, + DEVICE_TYPE, + ENABLE_FORWARD_USER_INFO_HEADERS, +) + + +router = APIRouter() + +# Constants +MAX_FILE_SIZE_MB = 25 +MAX_FILE_SIZE = MAX_FILE_SIZE_MB * 1024 * 1024 # Convert MB to bytes + +log = logging.getLogger(__name__) +log.setLevel(SRC_LOG_LEVELS["AUDIO"]) + +SPEECH_CACHE_DIR = Path(CACHE_DIR).joinpath("./audio/speech/") +SPEECH_CACHE_DIR.mkdir(parents=True, exist_ok=True) + + +########################################## +# +# Utility functions +# +########################################## + +from pydub import AudioSegment +from pydub.utils import mediainfo + + +def is_mp4_audio(file_path): + """Check if the given file is an MP4 audio file.""" + if not os.path.isfile(file_path): + print(f"File not found: {file_path}") + return False + + info = mediainfo(file_path) + if ( + info.get("codec_name") == "aac" + and info.get("codec_type") == "audio" + and info.get("codec_tag_string") == "mp4a" + ): + return True + return False + + +def convert_mp4_to_wav(file_path, output_path): + """Convert MP4 audio file to WAV format.""" + audio = AudioSegment.from_file(file_path, format="mp4") + audio.export(output_path, format="wav") + print(f"Converted {file_path} to {output_path}") + + +def set_faster_whisper_model(model: str, auto_update: bool = False): + whisper_model = None + if model: + from faster_whisper import WhisperModel + + faster_whisper_kwargs = { + "model_size_or_path": model, + "device": DEVICE_TYPE if DEVICE_TYPE and DEVICE_TYPE == "cuda" else "cpu", + "compute_type": "int8", + "download_root": WHISPER_MODEL_DIR, + "local_files_only": not auto_update, + } + + try: + whisper_model = WhisperModel(**faster_whisper_kwargs) + except Exception: + log.warning( + "WhisperModel initialization failed, attempting download with local_files_only=False" + ) + faster_whisper_kwargs["local_files_only"] = False + whisper_model = WhisperModel(**faster_whisper_kwargs) + return whisper_model + + +########################################## +# +# Audio API +# +########################################## + + +class TTSConfigForm(BaseModel): + OPENAI_API_BASE_URL: str + OPENAI_API_KEY: str + API_KEY: str + ENGINE: str + MODEL: str + VOICE: str + SPLIT_ON: str + AZURE_SPEECH_REGION: str + AZURE_SPEECH_OUTPUT_FORMAT: str + + +class STTConfigForm(BaseModel): + OPENAI_API_BASE_URL: str + OPENAI_API_KEY: str + ENGINE: str + MODEL: str + WHISPER_MODEL: str + + +class AudioConfigUpdateForm(BaseModel): + tts: TTSConfigForm + stt: STTConfigForm + + +@router.get("/config") +async def get_audio_config(request: Request, user=Depends(get_admin_user)): + return { + "tts": { + "OPENAI_API_BASE_URL": request.app.state.config.TTS_OPENAI_API_BASE_URL, + "OPENAI_API_KEY": request.app.state.config.TTS_OPENAI_API_KEY, + "API_KEY": request.app.state.config.TTS_API_KEY, + "ENGINE": request.app.state.config.TTS_ENGINE, + "MODEL": request.app.state.config.TTS_MODEL, + "VOICE": request.app.state.config.TTS_VOICE, + "SPLIT_ON": request.app.state.config.TTS_SPLIT_ON, + "AZURE_SPEECH_REGION": request.app.state.config.TTS_AZURE_SPEECH_REGION, + "AZURE_SPEECH_OUTPUT_FORMAT": request.app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT, + }, + "stt": { + "OPENAI_API_BASE_URL": request.app.state.config.STT_OPENAI_API_BASE_URL, + "OPENAI_API_KEY": request.app.state.config.STT_OPENAI_API_KEY, + "ENGINE": request.app.state.config.STT_ENGINE, + "MODEL": request.app.state.config.STT_MODEL, + "WHISPER_MODEL": request.app.state.config.WHISPER_MODEL, + }, + } + + +@router.post("/config/update") +async def update_audio_config( + request: Request, form_data: AudioConfigUpdateForm, user=Depends(get_admin_user) +): + request.app.state.config.TTS_OPENAI_API_BASE_URL = form_data.tts.OPENAI_API_BASE_URL + request.app.state.config.TTS_OPENAI_API_KEY = form_data.tts.OPENAI_API_KEY + request.app.state.config.TTS_API_KEY = form_data.tts.API_KEY + request.app.state.config.TTS_ENGINE = form_data.tts.ENGINE + request.app.state.config.TTS_MODEL = form_data.tts.MODEL + request.app.state.config.TTS_VOICE = form_data.tts.VOICE + request.app.state.config.TTS_SPLIT_ON = form_data.tts.SPLIT_ON + request.app.state.config.TTS_AZURE_SPEECH_REGION = form_data.tts.AZURE_SPEECH_REGION + request.app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT = ( + form_data.tts.AZURE_SPEECH_OUTPUT_FORMAT + ) + + request.app.state.config.STT_OPENAI_API_BASE_URL = form_data.stt.OPENAI_API_BASE_URL + request.app.state.config.STT_OPENAI_API_KEY = form_data.stt.OPENAI_API_KEY + request.app.state.config.STT_ENGINE = form_data.stt.ENGINE + request.app.state.config.STT_MODEL = form_data.stt.MODEL + request.app.state.config.WHISPER_MODEL = form_data.stt.WHISPER_MODEL + + if request.app.state.config.STT_ENGINE == "": + request.app.state.faster_whisper_model = set_faster_whisper_model( + form_data.stt.WHISPER_MODEL, WHISPER_MODEL_AUTO_UPDATE + ) + + return { + "tts": { + "OPENAI_API_BASE_URL": request.app.state.config.TTS_OPENAI_API_BASE_URL, + "OPENAI_API_KEY": request.app.state.config.TTS_OPENAI_API_KEY, + "API_KEY": request.app.state.config.TTS_API_KEY, + "ENGINE": request.app.state.config.TTS_ENGINE, + "MODEL": request.app.state.config.TTS_MODEL, + "VOICE": request.app.state.config.TTS_VOICE, + "SPLIT_ON": request.app.state.config.TTS_SPLIT_ON, + "AZURE_SPEECH_REGION": request.app.state.config.TTS_AZURE_SPEECH_REGION, + "AZURE_SPEECH_OUTPUT_FORMAT": request.app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT, + }, + "stt": { + "OPENAI_API_BASE_URL": request.app.state.config.STT_OPENAI_API_BASE_URL, + "OPENAI_API_KEY": request.app.state.config.STT_OPENAI_API_KEY, + "ENGINE": request.app.state.config.STT_ENGINE, + "MODEL": request.app.state.config.STT_MODEL, + "WHISPER_MODEL": request.app.state.config.WHISPER_MODEL, + }, + } + + +def load_speech_pipeline(): + from transformers import pipeline + from datasets import load_dataset + + if request.app.state.speech_synthesiser is None: + request.app.state.speech_synthesiser = pipeline( + "text-to-speech", "microsoft/speecht5_tts" + ) + + if request.app.state.speech_speaker_embeddings_dataset is None: + request.app.state.speech_speaker_embeddings_dataset = load_dataset( + "Matthijs/cmu-arctic-xvectors", split="validation" + ) + + +@router.post("/speech") +async def speech(request: Request, user=Depends(get_verified_user)): + body = await request.body() + name = hashlib.sha256(body).hexdigest() + + file_path = SPEECH_CACHE_DIR.joinpath(f"{name}.mp3") + file_body_path = SPEECH_CACHE_DIR.joinpath(f"{name}.json") + + # Check if the file already exists in the cache + if file_path.is_file(): + return FileResponse(file_path) + + payload = None + try: + payload = json.loads(body.decode("utf-8")) + except Exception as e: + log.exception(e) + raise HTTPException(status_code=400, detail="Invalid JSON payload") + + if request.app.state.config.TTS_ENGINE == "openai": + payload["model"] = request.app.state.config.TTS_MODEL + + try: + async with aiohttp.ClientSession() as session: + async with session.post( + url=f"{request.app.state.config.TTS_OPENAI_API_BASE_URL}/audio/speech", + data=payload, + headers={ + "Content-Type": "application/json", + "Authorization": f"Bearer {request.app.state.config.TTS_OPENAI_API_KEY}", + **( + { + "X-OpenWebUI-User-Name": user.name, + "X-OpenWebUI-User-Id": user.id, + "X-OpenWebUI-User-Email": user.email, + "X-OpenWebUI-User-Role": user.role, + } + if ENABLE_FORWARD_USER_INFO_HEADERS + else {} + ), + }, + ) as r: + r.raise_for_status() + + async with aiofiles.open(file_path, "wb") as f: + await f.write(await r.read()) + + async with aiofiles.open(file_body_path, "w") as f: + await f.write(json.dumps(json.loads(body.decode("utf-8")))) + + return FileResponse(file_path) + + except Exception as e: + log.exception(e) + detail = None + + try: + if r.status != 200: + res = await r.json() + if "error" in res: + detail = f"External: {res['error'].get('message', '')}" + except Exception: + detail = f"External: {e}" + + raise HTTPException( + status_code=getattr(r, "status", 500), + detail=detail if detail else "Open WebUI: Server Connection Error", + ) + + elif request.app.state.config.TTS_ENGINE == "elevenlabs": + voice_id = payload.get("voice", "") + + if voice_id not in get_available_voices(): + raise HTTPException( + status_code=400, + detail="Invalid voice id", + ) + + try: + async with aiohttp.ClientSession() as session: + async with session.post( + f"https://api.elevenlabs.io/v1/text-to-speech/{voice_id}", + json={ + "text": payload["input"], + "model_id": request.app.state.config.TTS_MODEL, + "voice_settings": {"stability": 0.5, "similarity_boost": 0.5}, + }, + headers={ + "Accept": "audio/mpeg", + "Content-Type": "application/json", + "xi-api-key": request.app.state.config.TTS_API_KEY, + }, + ) as r: + r.raise_for_status() + + async with aiofiles.open(file_path, "wb") as f: + await f.write(await r.read()) + + async with aiofiles.open(file_body_path, "w") as f: + await f.write(json.dumps(json.loads(body.decode("utf-8")))) + + return FileResponse(file_path) + + except Exception as e: + log.exception(e) + detail = None + + try: + if r.status != 200: + res = await r.json() + if "error" in res: + detail = f"External: {res['error'].get('message', '')}" + except Exception: + detail = f"External: {e}" + + raise HTTPException( + status_code=getattr(r, "status", 500), + detail=detail if detail else "Open WebUI: Server Connection Error", + ) + + elif request.app.state.config.TTS_ENGINE == "azure": + try: + payload = json.loads(body.decode("utf-8")) + except Exception as e: + log.exception(e) + raise HTTPException(status_code=400, detail="Invalid JSON payload") + + region = request.app.state.config.TTS_AZURE_SPEECH_REGION + language = request.app.state.config.TTS_VOICE + locale = "-".join(request.app.state.config.TTS_VOICE.split("-")[:1]) + output_format = request.app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT + + try: + data = f""" + {payload["input"]} + """ + async with aiohttp.ClientSession() as session: + async with session.post( + f"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1", + headers={ + "Ocp-Apim-Subscription-Key": request.app.state.config.TTS_API_KEY, + "Content-Type": "application/ssml+xml", + "X-Microsoft-OutputFormat": output_format, + }, + data=data, + ) as r: + r.raise_for_status() + + async with aiofiles.open(file_path, "wb") as f: + await f.write(await r.read()) + + return FileResponse(file_path) + + except Exception as e: + log.exception(e) + detail = None + + try: + if r.status != 200: + res = await r.json() + if "error" in res: + detail = f"External: {res['error'].get('message', '')}" + except Exception: + detail = f"External: {e}" + + raise HTTPException( + status_code=getattr(r, "status", 500), + detail=detail if detail else "Open WebUI: Server Connection Error", + ) + + elif request.app.state.config.TTS_ENGINE == "transformers": + payload = None + try: + payload = json.loads(body.decode("utf-8")) + except Exception as e: + log.exception(e) + raise HTTPException(status_code=400, detail="Invalid JSON payload") + + import torch + import soundfile as sf + + load_speech_pipeline() + + embeddings_dataset = request.app.state.speech_speaker_embeddings_dataset + + speaker_index = 6799 + try: + speaker_index = embeddings_dataset["filename"].index( + request.app.state.config.TTS_MODEL + ) + except Exception: + pass + + speaker_embedding = torch.tensor( + embeddings_dataset[speaker_index]["xvector"] + ).unsqueeze(0) + + speech = request.app.state.speech_synthesiser( + payload["input"], + forward_params={"speaker_embeddings": speaker_embedding}, + ) + + sf.write(file_path, speech["audio"], samplerate=speech["sampling_rate"]) + with open(file_body_path, "w") as f: + json.dump(json.loads(body.decode("utf-8")), f) + + return FileResponse(file_path) + + +def transcribe(request: Request, file_path): + print("transcribe", file_path) + filename = os.path.basename(file_path) + file_dir = os.path.dirname(file_path) + id = filename.split(".")[0] + + if request.app.state.config.STT_ENGINE == "": + if request.app.state.faster_whisper_model is None: + request.app.state.faster_whisper_model = set_faster_whisper_model( + request.app.state.config.WHISPER_MODEL + ) + + model = request.app.state.faster_whisper_model + segments, info = model.transcribe(file_path, beam_size=5) + log.info( + "Detected language '%s' with probability %f" + % (info.language, info.language_probability) + ) + + transcript = "".join([segment.text for segment in list(segments)]) + data = {"text": transcript.strip()} + + # save the transcript to a json file + transcript_file = f"{file_dir}/{id}.json" + with open(transcript_file, "w") as f: + json.dump(data, f) + + log.debug(data) + return data + elif request.app.state.config.STT_ENGINE == "openai": + if is_mp4_audio(file_path): + os.rename(file_path, file_path.replace(".wav", ".mp4")) + # Convert MP4 audio file to WAV format + convert_mp4_to_wav(file_path.replace(".wav", ".mp4"), file_path) + + r = None + try: + r = requests.post( + url=f"{request.app.state.config.STT_OPENAI_API_BASE_URL}/audio/transcriptions", + headers={ + "Authorization": f"Bearer {request.app.state.config.STT_OPENAI_API_KEY}" + }, + files={"file": (filename, open(file_path, "rb"))}, + data={"model": request.app.state.config.STT_MODEL}, + ) + + r.raise_for_status() + data = r.json() + + # save the transcript to a json file + transcript_file = f"{file_dir}/{id}.json" + with open(transcript_file, "w") as f: + json.dump(data, f) + + return data + except Exception as e: + log.exception(e) + + detail = None + if r is not None: + try: + res = r.json() + if "error" in res: + detail = f"External: {res['error'].get('message', '')}" + except Exception: + detail = f"External: {e}" + + raise Exception(detail if detail else "Open WebUI: Server Connection Error") + + +def compress_audio(file_path): + if os.path.getsize(file_path) > MAX_FILE_SIZE: + file_dir = os.path.dirname(file_path) + audio = AudioSegment.from_file(file_path) + audio = audio.set_frame_rate(16000).set_channels(1) # Compress audio + compressed_path = f"{file_dir}/{id}_compressed.opus" + audio.export(compressed_path, format="opus", bitrate="32k") + log.debug(f"Compressed audio to {compressed_path}") + + if ( + os.path.getsize(compressed_path) > MAX_FILE_SIZE + ): # Still larger than MAX_FILE_SIZE after compression + raise Exception(ERROR_MESSAGES.FILE_TOO_LARGE(size=f"{MAX_FILE_SIZE_MB}MB")) + return compressed_path + else: + return file_path + + +@router.post("/transcriptions") +def transcription( + request: Request, + file: UploadFile = File(...), + user=Depends(get_verified_user), +): + log.info(f"file.content_type: {file.content_type}") + + if file.content_type not in ["audio/mpeg", "audio/wav", "audio/ogg", "audio/x-m4a"]: + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=ERROR_MESSAGES.FILE_NOT_SUPPORTED, + ) + + try: + ext = file.filename.split(".")[-1] + id = uuid.uuid4() + + filename = f"{id}.{ext}" + contents = file.file.read() + + file_dir = f"{CACHE_DIR}/audio/transcriptions" + os.makedirs(file_dir, exist_ok=True) + file_path = f"{file_dir}/{filename}" + + with open(file_path, "wb") as f: + f.write(contents) + + try: + try: + file_path = compress_audio(file_path) + except Exception as e: + log.exception(e) + + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=ERROR_MESSAGES.DEFAULT(e), + ) + + data = transcribe(request, file_path) + file_path = file_path.split("/")[-1] + return {**data, "filename": file_path} + except Exception as e: + log.exception(e) + + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=ERROR_MESSAGES.DEFAULT(e), + ) + + except Exception as e: + log.exception(e) + + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=ERROR_MESSAGES.DEFAULT(e), + ) + + +def get_available_models(request: Request) -> list[dict]: + available_models = [] + if request.app.state.config.TTS_ENGINE == "openai": + available_models = [{"id": "tts-1"}, {"id": "tts-1-hd"}] + elif request.app.state.config.TTS_ENGINE == "elevenlabs": + try: + response = requests.get( + "https://api.elevenlabs.io/v1/models", + headers={ + "xi-api-key": request.app.state.config.TTS_API_KEY, + "Content-Type": "application/json", + }, + timeout=5, + ) + response.raise_for_status() + models = response.json() + + available_models = [ + {"name": model["name"], "id": model["model_id"]} for model in models + ] + except requests.RequestException as e: + log.error(f"Error fetching voices: {str(e)}") + return available_models + + +@router.get("/models") +async def get_models(request: Request, user=Depends(get_verified_user)): + return {"models": get_available_models(request)} + + +def get_available_voices(request) -> dict: + """Returns {voice_id: voice_name} dict""" + available_voices = {} + if request.app.state.config.TTS_ENGINE == "openai": + available_voices = { + "alloy": "alloy", + "echo": "echo", + "fable": "fable", + "onyx": "onyx", + "nova": "nova", + "shimmer": "shimmer", + } + elif request.app.state.config.TTS_ENGINE == "elevenlabs": + try: + available_voices = get_elevenlabs_voices( + api_key=request.app.state.config.TTS_API_KEY + ) + except Exception: + # Avoided @lru_cache with exception + pass + elif request.app.state.config.TTS_ENGINE == "azure": + try: + region = request.app.state.config.TTS_AZURE_SPEECH_REGION + url = f"https://{region}.tts.speech.microsoft.com/cognitiveservices/voices/list" + headers = { + "Ocp-Apim-Subscription-Key": request.app.state.config.TTS_API_KEY + } + + response = requests.get(url, headers=headers) + response.raise_for_status() + voices = response.json() + + for voice in voices: + available_voices[voice["ShortName"]] = ( + f"{voice['DisplayName']} ({voice['ShortName']})" + ) + except requests.RequestException as e: + log.error(f"Error fetching voices: {str(e)}") + + return available_voices + + +@lru_cache +def get_elevenlabs_voices(api_key: str) -> dict: + """ + Note, set the following in your .env file to use Elevenlabs: + AUDIO_TTS_ENGINE=elevenlabs + AUDIO_TTS_API_KEY=sk_... # Your Elevenlabs API key + AUDIO_TTS_VOICE=EXAVITQu4vr4xnSDxMaL # From https://api.elevenlabs.io/v1/voices + AUDIO_TTS_MODEL=eleven_multilingual_v2 + """ + + try: + # TODO: Add retries + response = requests.get( + "https://api.elevenlabs.io/v1/voices", + headers={ + "xi-api-key": api_key, + "Content-Type": "application/json", + }, + ) + response.raise_for_status() + voices_data = response.json() + + voices = {} + for voice in voices_data.get("voices", []): + voices[voice["voice_id"]] = voice["name"] + except requests.RequestException as e: + # Avoid @lru_cache with exception + log.error(f"Error fetching voices: {str(e)}") + raise RuntimeError(f"Error fetching voices: {str(e)}") + + return voices + + +@router.get("/voices") +async def get_voices(request: Request, user=Depends(get_verified_user)): + return { + "voices": [ + {"id": k, "name": v} for k, v in get_available_voices(request).items() + ] + } diff --git a/backend/open_webui/apps/webui/routers/auths.py b/backend/open_webui/routers/auths.py similarity index 99% rename from backend/open_webui/apps/webui/routers/auths.py rename to backend/open_webui/routers/auths.py index 094ce568f..0b1f42edf 100644 --- a/backend/open_webui/apps/webui/routers/auths.py +++ b/backend/open_webui/routers/auths.py @@ -5,7 +5,7 @@ import datetime import logging from aiohttp import ClientSession -from open_webui.apps.webui.models.auths import ( +from open_webui.models.auths import ( AddUserForm, ApiKey, Auths, @@ -18,7 +18,7 @@ from open_webui.apps.webui.models.auths import ( UpdateProfileForm, UserResponse, ) -from open_webui.apps.webui.models.users import Users +from open_webui.models.users import Users from open_webui.constants import ERROR_MESSAGES, WEBHOOK_MESSAGES from open_webui.env import ( diff --git a/backend/open_webui/apps/webui/routers/chats.py b/backend/open_webui/routers/chats.py similarity index 99% rename from backend/open_webui/apps/webui/routers/chats.py rename to backend/open_webui/routers/chats.py index ec5dae4bf..5e0e75e24 100644 --- a/backend/open_webui/apps/webui/routers/chats.py +++ b/backend/open_webui/routers/chats.py @@ -2,15 +2,15 @@ import json import logging from typing import Optional -from open_webui.apps.webui.models.chats import ( +from open_webui.models.chats import ( ChatForm, ChatImportForm, ChatResponse, Chats, ChatTitleIdResponse, ) -from open_webui.apps.webui.models.tags import TagModel, Tags -from open_webui.apps.webui.models.folders import Folders +from open_webui.models.tags import TagModel, Tags +from open_webui.models.folders import Folders from open_webui.config import ENABLE_ADMIN_CHAT_ACCESS, ENABLE_ADMIN_EXPORT from open_webui.constants import ERROR_MESSAGES diff --git a/backend/open_webui/apps/webui/routers/configs.py b/backend/open_webui/routers/configs.py similarity index 100% rename from backend/open_webui/apps/webui/routers/configs.py rename to backend/open_webui/routers/configs.py diff --git a/backend/open_webui/apps/webui/routers/evaluations.py b/backend/open_webui/routers/evaluations.py similarity index 97% rename from backend/open_webui/apps/webui/routers/evaluations.py rename to backend/open_webui/routers/evaluations.py index 0bcee2a79..f0c4a6b06 100644 --- a/backend/open_webui/apps/webui/routers/evaluations.py +++ b/backend/open_webui/routers/evaluations.py @@ -2,8 +2,8 @@ from typing import Optional from fastapi import APIRouter, Depends, HTTPException, status, Request from pydantic import BaseModel -from open_webui.apps.webui.models.users import Users, UserModel -from open_webui.apps.webui.models.feedbacks import ( +from open_webui.models.users import Users, UserModel +from open_webui.models.feedbacks import ( FeedbackModel, FeedbackResponse, FeedbackForm, diff --git a/backend/open_webui/apps/webui/routers/files.py b/backend/open_webui/routers/files.py similarity index 95% rename from backend/open_webui/apps/webui/routers/files.py rename to backend/open_webui/routers/files.py index 4b7cf1ed4..e56eef273 100644 --- a/backend/open_webui/apps/webui/routers/files.py +++ b/backend/open_webui/routers/files.py @@ -8,20 +8,20 @@ import mimetypes from open_webui.storage.provider import Storage -from open_webui.apps.webui.models.files import ( +from open_webui.models.files import ( FileForm, FileModel, FileModelResponse, Files, ) -from open_webui.apps.retrieval.main import process_file, ProcessFileForm +from open_webui.routers.retrieval import process_file, ProcessFileForm from open_webui.config import UPLOAD_DIR from open_webui.env import SRC_LOG_LEVELS from open_webui.constants import ERROR_MESSAGES -from fastapi import APIRouter, Depends, File, HTTPException, UploadFile, status +from fastapi import APIRouter, Depends, File, HTTPException, UploadFile, status, Request from fastapi.responses import FileResponse, StreamingResponse @@ -39,7 +39,9 @@ router = APIRouter() @router.post("/", response_model=FileModelResponse) -def upload_file(file: UploadFile = File(...), user=Depends(get_verified_user)): +def upload_file( + request: Request, file: UploadFile = File(...), user=Depends(get_verified_user) +): log.info(f"file.content_type: {file.content_type}") try: unsanitized_filename = file.filename @@ -68,7 +70,7 @@ def upload_file(file: UploadFile = File(...), user=Depends(get_verified_user)): ) try: - process_file(ProcessFileForm(file_id=id)) + process_file(request, ProcessFileForm(file_id=id)) file_item = Files.get_file_by_id(id=id) except Exception as e: log.exception(e) @@ -183,13 +185,15 @@ class ContentForm(BaseModel): @router.post("/{id}/data/content/update") async def update_file_data_content_by_id( - id: str, form_data: ContentForm, user=Depends(get_verified_user) + request: Request, id: str, form_data: ContentForm, user=Depends(get_verified_user) ): file = Files.get_file_by_id(id) if file and (file.user_id == user.id or user.role == "admin"): try: - process_file(ProcessFileForm(file_id=id, content=form_data.content)) + process_file( + request, ProcessFileForm(file_id=id, content=form_data.content) + ) file = Files.get_file_by_id(id=id) except Exception as e: log.exception(e) diff --git a/backend/open_webui/apps/webui/routers/folders.py b/backend/open_webui/routers/folders.py similarity index 98% rename from backend/open_webui/apps/webui/routers/folders.py rename to backend/open_webui/routers/folders.py index f05781476..ca2fbd213 100644 --- a/backend/open_webui/apps/webui/routers/folders.py +++ b/backend/open_webui/routers/folders.py @@ -8,12 +8,12 @@ from pydantic import BaseModel import mimetypes -from open_webui.apps.webui.models.folders import ( +from open_webui.models.folders import ( FolderForm, FolderModel, Folders, ) -from open_webui.apps.webui.models.chats import Chats +from open_webui.models.chats import Chats from open_webui.config import UPLOAD_DIR from open_webui.env import SRC_LOG_LEVELS diff --git a/backend/open_webui/apps/webui/routers/functions.py b/backend/open_webui/routers/functions.py similarity index 98% rename from backend/open_webui/apps/webui/routers/functions.py rename to backend/open_webui/routers/functions.py index bdd422b95..7f3305f25 100644 --- a/backend/open_webui/apps/webui/routers/functions.py +++ b/backend/open_webui/routers/functions.py @@ -2,13 +2,13 @@ import os from pathlib import Path from typing import Optional -from open_webui.apps.webui.models.functions import ( +from open_webui.models.functions import ( FunctionForm, FunctionModel, FunctionResponse, Functions, ) -from open_webui.apps.webui.utils import load_function_module_by_id, replace_imports +from open_webui.utils.plugin import load_function_module_by_id, replace_imports from open_webui.config import CACHE_DIR from open_webui.constants import ERROR_MESSAGES from fastapi import APIRouter, Depends, HTTPException, Request, status diff --git a/backend/open_webui/apps/webui/routers/groups.py b/backend/open_webui/routers/groups.py similarity index 98% rename from backend/open_webui/apps/webui/routers/groups.py rename to backend/open_webui/routers/groups.py index ef392fb6a..e8f8994a4 100644 --- a/backend/open_webui/apps/webui/routers/groups.py +++ b/backend/open_webui/routers/groups.py @@ -2,7 +2,7 @@ import os from pathlib import Path from typing import Optional -from open_webui.apps.webui.models.groups import ( +from open_webui.models.groups import ( Groups, GroupForm, GroupUpdateForm, diff --git a/backend/open_webui/apps/images/main.py b/backend/open_webui/routers/images.py similarity index 58% rename from backend/open_webui/apps/images/main.py rename to backend/open_webui/routers/images.py index 14209df2f..3f51fbdb4 100644 --- a/backend/open_webui/apps/images/main.py +++ b/backend/open_webui/routers/images.py @@ -9,38 +9,24 @@ from pathlib import Path from typing import Optional import requests -from open_webui.apps.images.utils.comfyui import ( + + +from fastapi import Depends, FastAPI, HTTPException, Request, APIRouter +from fastapi.middleware.cors import CORSMiddleware +from pydantic import BaseModel + + +from open_webui.config import CACHE_DIR +from open_webui.constants import ERROR_MESSAGES +from open_webui.env import ENV, SRC_LOG_LEVELS, ENABLE_FORWARD_USER_INFO_HEADERS + +from open_webui.utils.auth import get_admin_user, get_verified_user +from open_webui.utils.images.comfyui import ( ComfyUIGenerateImageForm, ComfyUIWorkflow, comfyui_generate_image, ) -from open_webui.config import ( - AUTOMATIC1111_API_AUTH, - AUTOMATIC1111_BASE_URL, - AUTOMATIC1111_CFG_SCALE, - AUTOMATIC1111_SAMPLER, - AUTOMATIC1111_SCHEDULER, - CACHE_DIR, - COMFYUI_BASE_URL, - COMFYUI_WORKFLOW, - COMFYUI_WORKFLOW_NODES, - CORS_ALLOW_ORIGIN, - ENABLE_IMAGE_GENERATION, - IMAGE_GENERATION_ENGINE, - IMAGE_GENERATION_MODEL, - IMAGE_SIZE, - IMAGE_STEPS, - IMAGES_OPENAI_API_BASE_URL, - IMAGES_OPENAI_API_KEY, - AppConfig, -) -from open_webui.constants import ERROR_MESSAGES -from open_webui.env import ENV, SRC_LOG_LEVELS, ENABLE_FORWARD_USER_INFO_HEADERS -from fastapi import Depends, FastAPI, HTTPException, Request -from fastapi.middleware.cors import CORSMiddleware -from pydantic import BaseModel -from open_webui.utils.auth import get_admin_user, get_verified_user log = logging.getLogger(__name__) log.setLevel(SRC_LOG_LEVELS["IMAGES"]) @@ -48,63 +34,30 @@ log.setLevel(SRC_LOG_LEVELS["IMAGES"]) IMAGE_CACHE_DIR = Path(CACHE_DIR).joinpath("./image/generations/") IMAGE_CACHE_DIR.mkdir(parents=True, exist_ok=True) -app = FastAPI( - docs_url="/docs" if ENV == "dev" else None, - openapi_url="/openapi.json" if ENV == "dev" else None, - redoc_url=None, -) -app.add_middleware( - CORSMiddleware, - allow_origins=CORS_ALLOW_ORIGIN, - allow_credentials=True, - allow_methods=["*"], - allow_headers=["*"], -) - -app.state.config = AppConfig() - -app.state.config.ENGINE = IMAGE_GENERATION_ENGINE -app.state.config.ENABLED = ENABLE_IMAGE_GENERATION - -app.state.config.OPENAI_API_BASE_URL = IMAGES_OPENAI_API_BASE_URL -app.state.config.OPENAI_API_KEY = IMAGES_OPENAI_API_KEY - -app.state.config.MODEL = IMAGE_GENERATION_MODEL - -app.state.config.AUTOMATIC1111_BASE_URL = AUTOMATIC1111_BASE_URL -app.state.config.AUTOMATIC1111_API_AUTH = AUTOMATIC1111_API_AUTH -app.state.config.AUTOMATIC1111_CFG_SCALE = AUTOMATIC1111_CFG_SCALE -app.state.config.AUTOMATIC1111_SAMPLER = AUTOMATIC1111_SAMPLER -app.state.config.AUTOMATIC1111_SCHEDULER = AUTOMATIC1111_SCHEDULER -app.state.config.COMFYUI_BASE_URL = COMFYUI_BASE_URL -app.state.config.COMFYUI_WORKFLOW = COMFYUI_WORKFLOW -app.state.config.COMFYUI_WORKFLOW_NODES = COMFYUI_WORKFLOW_NODES - -app.state.config.IMAGE_SIZE = IMAGE_SIZE -app.state.config.IMAGE_STEPS = IMAGE_STEPS +router = APIRouter() -@app.get("/config") +@router.get("/config") async def get_config(request: Request, user=Depends(get_admin_user)): return { - "enabled": app.state.config.ENABLED, - "engine": app.state.config.ENGINE, + "enabled": request.app.state.config.ENABLE_IMAGE_GENERATION, + "engine": request.app.state.config.IMAGE_GENERATION_ENGINE, "openai": { - "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL, - "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY, + "OPENAI_API_BASE_URL": request.app.state.config.IMAGES_OPENAI_API_BASE_URL, + "OPENAI_API_KEY": request.app.state.config.IMAGES_OPENAI_API_KEY, }, "automatic1111": { - "AUTOMATIC1111_BASE_URL": app.state.config.AUTOMATIC1111_BASE_URL, - "AUTOMATIC1111_API_AUTH": app.state.config.AUTOMATIC1111_API_AUTH, - "AUTOMATIC1111_CFG_SCALE": app.state.config.AUTOMATIC1111_CFG_SCALE, - "AUTOMATIC1111_SAMPLER": app.state.config.AUTOMATIC1111_SAMPLER, - "AUTOMATIC1111_SCHEDULER": app.state.config.AUTOMATIC1111_SCHEDULER, + "AUTOMATIC1111_BASE_URL": request.app.state.config.AUTOMATIC1111_BASE_URL, + "AUTOMATIC1111_API_AUTH": request.app.state.config.AUTOMATIC1111_API_AUTH, + "AUTOMATIC1111_CFG_SCALE": request.app.state.config.AUTOMATIC1111_CFG_SCALE, + "AUTOMATIC1111_SAMPLER": request.app.state.config.AUTOMATIC1111_SAMPLER, + "AUTOMATIC1111_SCHEDULER": request.app.state.config.AUTOMATIC1111_SCHEDULER, }, "comfyui": { - "COMFYUI_BASE_URL": app.state.config.COMFYUI_BASE_URL, - "COMFYUI_WORKFLOW": app.state.config.COMFYUI_WORKFLOW, - "COMFYUI_WORKFLOW_NODES": app.state.config.COMFYUI_WORKFLOW_NODES, + "COMFYUI_BASE_URL": request.app.state.config.COMFYUI_BASE_URL, + "COMFYUI_WORKFLOW": request.app.state.config.COMFYUI_WORKFLOW, + "COMFYUI_WORKFLOW_NODES": request.app.state.config.COMFYUI_WORKFLOW_NODES, }, } @@ -136,133 +89,156 @@ class ConfigForm(BaseModel): comfyui: ComfyUIConfigForm -@app.post("/config/update") -async def update_config(form_data: ConfigForm, user=Depends(get_admin_user)): - app.state.config.ENGINE = form_data.engine - app.state.config.ENABLED = form_data.enabled +@router.post("/config/update") +async def update_config( + request: Request, form_data: ConfigForm, user=Depends(get_admin_user) +): + request.app.state.config.IMAGE_GENERATION_ENGINE = form_data.engine + request.app.state.config.ENABLE_IMAGE_GENERATION = form_data.enabled - app.state.config.OPENAI_API_BASE_URL = form_data.openai.OPENAI_API_BASE_URL - app.state.config.OPENAI_API_KEY = form_data.openai.OPENAI_API_KEY + request.app.state.config.IMAGES_OPENAI_API_BASE_URL = ( + form_data.openai.OPENAI_API_BASE_URL + ) + request.app.state.config.IMAGES_OPENAI_API_KEY = form_data.openai.OPENAI_API_KEY - app.state.config.AUTOMATIC1111_BASE_URL = ( + request.app.state.config.AUTOMATIC1111_BASE_URL = ( form_data.automatic1111.AUTOMATIC1111_BASE_URL ) - app.state.config.AUTOMATIC1111_API_AUTH = ( + request.app.state.config.AUTOMATIC1111_API_AUTH = ( form_data.automatic1111.AUTOMATIC1111_API_AUTH ) - app.state.config.AUTOMATIC1111_CFG_SCALE = ( + request.app.state.config.AUTOMATIC1111_CFG_SCALE = ( float(form_data.automatic1111.AUTOMATIC1111_CFG_SCALE) if form_data.automatic1111.AUTOMATIC1111_CFG_SCALE else None ) - app.state.config.AUTOMATIC1111_SAMPLER = ( + request.app.state.config.AUTOMATIC1111_SAMPLER = ( form_data.automatic1111.AUTOMATIC1111_SAMPLER if form_data.automatic1111.AUTOMATIC1111_SAMPLER else None ) - app.state.config.AUTOMATIC1111_SCHEDULER = ( + request.app.state.config.AUTOMATIC1111_SCHEDULER = ( form_data.automatic1111.AUTOMATIC1111_SCHEDULER if form_data.automatic1111.AUTOMATIC1111_SCHEDULER else None ) - app.state.config.COMFYUI_BASE_URL = form_data.comfyui.COMFYUI_BASE_URL.strip("/") - app.state.config.COMFYUI_WORKFLOW = form_data.comfyui.COMFYUI_WORKFLOW - app.state.config.COMFYUI_WORKFLOW_NODES = form_data.comfyui.COMFYUI_WORKFLOW_NODES + request.app.state.config.COMFYUI_BASE_URL = ( + form_data.comfyui.COMFYUI_BASE_URL.strip("/") + ) + request.app.state.config.COMFYUI_WORKFLOW = form_data.comfyui.COMFYUI_WORKFLOW + request.app.state.config.COMFYUI_WORKFLOW_NODES = ( + form_data.comfyui.COMFYUI_WORKFLOW_NODES + ) return { - "enabled": app.state.config.ENABLED, - "engine": app.state.config.ENGINE, + "enabled": request.app.state.config.ENABLE_IMAGE_GENERATION, + "engine": request.app.state.config.IMAGE_GENERATION_ENGINE, "openai": { - "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL, - "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY, + "OPENAI_API_BASE_URL": request.app.state.config.IMAGES_OPENAI_API_BASE_URL, + "OPENAI_API_KEY": request.app.state.config.IMAGES_OPENAI_API_KEY, }, "automatic1111": { - "AUTOMATIC1111_BASE_URL": app.state.config.AUTOMATIC1111_BASE_URL, - "AUTOMATIC1111_API_AUTH": app.state.config.AUTOMATIC1111_API_AUTH, - "AUTOMATIC1111_CFG_SCALE": app.state.config.AUTOMATIC1111_CFG_SCALE, - "AUTOMATIC1111_SAMPLER": app.state.config.AUTOMATIC1111_SAMPLER, - "AUTOMATIC1111_SCHEDULER": app.state.config.AUTOMATIC1111_SCHEDULER, + "AUTOMATIC1111_BASE_URL": request.app.state.config.AUTOMATIC1111_BASE_URL, + "AUTOMATIC1111_API_AUTH": request.app.state.config.AUTOMATIC1111_API_AUTH, + "AUTOMATIC1111_CFG_SCALE": request.app.state.config.AUTOMATIC1111_CFG_SCALE, + "AUTOMATIC1111_SAMPLER": request.app.state.config.AUTOMATIC1111_SAMPLER, + "AUTOMATIC1111_SCHEDULER": request.app.state.config.AUTOMATIC1111_SCHEDULER, }, "comfyui": { - "COMFYUI_BASE_URL": app.state.config.COMFYUI_BASE_URL, - "COMFYUI_WORKFLOW": app.state.config.COMFYUI_WORKFLOW, - "COMFYUI_WORKFLOW_NODES": app.state.config.COMFYUI_WORKFLOW_NODES, + "COMFYUI_BASE_URL": request.app.state.config.COMFYUI_BASE_URL, + "COMFYUI_WORKFLOW": request.app.state.config.COMFYUI_WORKFLOW, + "COMFYUI_WORKFLOW_NODES": request.app.state.config.COMFYUI_WORKFLOW_NODES, }, } -def get_automatic1111_api_auth(): - if app.state.config.AUTOMATIC1111_API_AUTH is None: +def get_automatic1111_api_auth(request: Request): + if request.app.state.config.AUTOMATIC1111_API_AUTH is None: return "" else: - auth1111_byte_string = app.state.config.AUTOMATIC1111_API_AUTH.encode("utf-8") + auth1111_byte_string = request.app.state.config.AUTOMATIC1111_API_AUTH.encode( + "utf-8" + ) auth1111_base64_encoded_bytes = base64.b64encode(auth1111_byte_string) auth1111_base64_encoded_string = auth1111_base64_encoded_bytes.decode("utf-8") return f"Basic {auth1111_base64_encoded_string}" -@app.get("/config/url/verify") -async def verify_url(user=Depends(get_admin_user)): - if app.state.config.ENGINE == "automatic1111": +@router.get("/config/url/verify") +async def verify_url(request: Request, user=Depends(get_admin_user)): + if request.app.state.config.IMAGE_GENERATION_ENGINE == "automatic1111": try: r = requests.get( - url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options", - headers={"authorization": get_automatic1111_api_auth()}, + url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options", + headers={"authorization": get_automatic1111_api_auth(request)}, ) r.raise_for_status() return True except Exception: - app.state.config.ENABLED = False + request.app.state.config.ENABLE_IMAGE_GENERATION = False raise HTTPException(status_code=400, detail=ERROR_MESSAGES.INVALID_URL) - elif app.state.config.ENGINE == "comfyui": + elif request.app.state.config.IMAGE_GENERATION_ENGINE == "comfyui": try: - r = requests.get(url=f"{app.state.config.COMFYUI_BASE_URL}/object_info") + r = requests.get( + url=f"{request.app.state.config.COMFYUI_BASE_URL}/object_info" + ) r.raise_for_status() return True except Exception: - app.state.config.ENABLED = False + request.app.state.config.ENABLE_IMAGE_GENERATION = False raise HTTPException(status_code=400, detail=ERROR_MESSAGES.INVALID_URL) else: return True -def set_image_model(model: str): +def set_image_model(request: Request, model: str): log.info(f"Setting image model to {model}") - app.state.config.MODEL = model - if app.state.config.ENGINE in ["", "automatic1111"]: + request.app.state.config.IMAGE_GENERATION_MODEL = model + if request.app.state.config.IMAGE_GENERATION_ENGINE in ["", "automatic1111"]: api_auth = get_automatic1111_api_auth() r = requests.get( - url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options", + url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options", headers={"authorization": api_auth}, ) options = r.json() if model != options["sd_model_checkpoint"]: options["sd_model_checkpoint"] = model r = requests.post( - url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options", + url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options", json=options, headers={"authorization": api_auth}, ) - return app.state.config.MODEL + return request.app.state.config.IMAGE_GENERATION_MODEL -def get_image_model(): - if app.state.config.ENGINE == "openai": - return app.state.config.MODEL if app.state.config.MODEL else "dall-e-2" - elif app.state.config.ENGINE == "comfyui": - return app.state.config.MODEL if app.state.config.MODEL else "" - elif app.state.config.ENGINE == "automatic1111" or app.state.config.ENGINE == "": +def get_image_model(request): + if request.app.state.config.IMAGE_GENERATION_ENGINE == "openai": + return ( + request.app.state.config.IMAGE_GENERATION_MODEL + if request.app.state.config.IMAGE_GENERATION_MODEL + else "dall-e-2" + ) + elif request.app.state.config.IMAGE_GENERATION_ENGINE == "comfyui": + return ( + request.app.state.config.IMAGE_GENERATION_MODEL + if request.app.state.config.IMAGE_GENERATION_MODEL + else "" + ) + elif ( + request.app.state.config.IMAGE_GENERATION_ENGINE == "automatic1111" + or request.app.state.config.IMAGE_GENERATION_ENGINE == "" + ): try: r = requests.get( - url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options", + url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options", headers={"authorization": get_automatic1111_api_auth()}, ) options = r.json() return options["sd_model_checkpoint"] except Exception as e: - app.state.config.ENABLED = False + request.app.state.config.ENABLE_IMAGE_GENERATION = False raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e)) @@ -272,23 +248,25 @@ class ImageConfigForm(BaseModel): IMAGE_STEPS: int -@app.get("/image/config") -async def get_image_config(user=Depends(get_admin_user)): +@router.get("/image/config") +async def get_image_config(request: Request, user=Depends(get_admin_user)): return { - "MODEL": app.state.config.MODEL, - "IMAGE_SIZE": app.state.config.IMAGE_SIZE, - "IMAGE_STEPS": app.state.config.IMAGE_STEPS, + "MODEL": request.app.state.config.IMAGE_GENERATION_MODEL, + "IMAGE_SIZE": request.app.state.config.IMAGE_SIZE, + "IMAGE_STEPS": request.app.state.config.IMAGE_STEPS, } -@app.post("/image/config/update") -async def update_image_config(form_data: ImageConfigForm, user=Depends(get_admin_user)): +@router.post("/image/config/update") +async def update_image_config( + request: Request, form_data: ImageConfigForm, user=Depends(get_admin_user) +): - set_image_model(form_data.MODEL) + set_image_model(request, form_data.MODEL) pattern = r"^\d+x\d+$" if re.match(pattern, form_data.IMAGE_SIZE): - app.state.config.IMAGE_SIZE = form_data.IMAGE_SIZE + request.app.state.config.IMAGE_SIZE = form_data.IMAGE_SIZE else: raise HTTPException( status_code=400, @@ -296,7 +274,7 @@ async def update_image_config(form_data: ImageConfigForm, user=Depends(get_admin ) if form_data.IMAGE_STEPS >= 0: - app.state.config.IMAGE_STEPS = form_data.IMAGE_STEPS + request.app.state.config.IMAGE_STEPS = form_data.IMAGE_STEPS else: raise HTTPException( status_code=400, @@ -304,29 +282,31 @@ async def update_image_config(form_data: ImageConfigForm, user=Depends(get_admin ) return { - "MODEL": app.state.config.MODEL, - "IMAGE_SIZE": app.state.config.IMAGE_SIZE, - "IMAGE_STEPS": app.state.config.IMAGE_STEPS, + "MODEL": request.app.state.config.IMAGE_GENERATION_MODEL, + "IMAGE_SIZE": request.app.state.config.IMAGE_SIZE, + "IMAGE_STEPS": request.app.state.config.IMAGE_STEPS, } -@app.get("/models") -def get_models(user=Depends(get_verified_user)): +@router.get("/models") +def get_models(request: Request, user=Depends(get_verified_user)): try: - if app.state.config.ENGINE == "openai": + if request.app.state.config.IMAGE_GENERATION_ENGINE == "openai": return [ {"id": "dall-e-2", "name": "DALL·E 2"}, {"id": "dall-e-3", "name": "DALL·E 3"}, ] - elif app.state.config.ENGINE == "comfyui": + elif request.app.state.config.IMAGE_GENERATION_ENGINE == "comfyui": # TODO - get models from comfyui - r = requests.get(url=f"{app.state.config.COMFYUI_BASE_URL}/object_info") + r = requests.get( + url=f"{request.app.state.config.COMFYUI_BASE_URL}/object_info" + ) info = r.json() - workflow = json.loads(app.state.config.COMFYUI_WORKFLOW) + workflow = json.loads(request.app.state.config.COMFYUI_WORKFLOW) model_node_id = None - for node in app.state.config.COMFYUI_WORKFLOW_NODES: + for node in request.app.state.config.COMFYUI_WORKFLOW_NODES: if node["type"] == "model": if node["node_ids"]: model_node_id = node["node_ids"][0] @@ -362,10 +342,11 @@ def get_models(user=Depends(get_verified_user)): ) ) elif ( - app.state.config.ENGINE == "automatic1111" or app.state.config.ENGINE == "" + request.app.state.config.IMAGE_GENERATION_ENGINE == "automatic1111" + or request.app.state.config.IMAGE_GENERATION_ENGINE == "" ): r = requests.get( - url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/sd-models", + url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/sd-models", headers={"authorization": get_automatic1111_api_auth()}, ) models = r.json() @@ -376,7 +357,7 @@ def get_models(user=Depends(get_verified_user)): ) ) except Exception as e: - app.state.config.ENABLED = False + request.app.state.config.ENABLE_IMAGE_GENERATION = False raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e)) @@ -448,18 +429,21 @@ def save_url_image(url): return None -@app.post("/generations") +@router.post("/generations") async def image_generations( + request: Request, form_data: GenerateImageForm, user=Depends(get_verified_user), ): - width, height = tuple(map(int, app.state.config.IMAGE_SIZE.split("x"))) + width, height = tuple(map(int, request.app.state.config.IMAGE_SIZE.split("x"))) r = None try: - if app.state.config.ENGINE == "openai": + if request.app.state.config.IMAGE_GENERATION_ENGINE == "openai": headers = {} - headers["Authorization"] = f"Bearer {app.state.config.OPENAI_API_KEY}" + headers["Authorization"] = ( + f"Bearer {request.app.state.config.IMAGES_OPENAI_API_KEY}" + ) headers["Content-Type"] = "application/json" if ENABLE_FORWARD_USER_INFO_HEADERS: @@ -470,14 +454,16 @@ async def image_generations( data = { "model": ( - app.state.config.MODEL - if app.state.config.MODEL != "" + request.app.state.config.IMAGE_GENERATION_MODEL + if request.app.state.config.IMAGE_GENERATION_MODEL != "" else "dall-e-2" ), "prompt": form_data.prompt, "n": form_data.n, "size": ( - form_data.size if form_data.size else app.state.config.IMAGE_SIZE + form_data.size + if form_data.size + else request.app.state.config.IMAGE_SIZE ), "response_format": "b64_json", } @@ -485,7 +471,7 @@ async def image_generations( # Use asyncio.to_thread for the requests.post call r = await asyncio.to_thread( requests.post, - url=f"{app.state.config.OPENAI_API_BASE_URL}/images/generations", + url=f"{request.app.state.config.IMAGES_OPENAI_API_BASE_URL}/images/generations", json=data, headers=headers, ) @@ -505,7 +491,7 @@ async def image_generations( return images - elif app.state.config.ENGINE == "comfyui": + elif request.app.state.config.IMAGE_GENERATION_ENGINE == "comfyui": data = { "prompt": form_data.prompt, "width": width, @@ -513,8 +499,8 @@ async def image_generations( "n": form_data.n, } - if app.state.config.IMAGE_STEPS is not None: - data["steps"] = app.state.config.IMAGE_STEPS + if request.app.state.config.IMAGE_STEPS is not None: + data["steps"] = request.app.state.config.IMAGE_STEPS if form_data.negative_prompt is not None: data["negative_prompt"] = form_data.negative_prompt @@ -523,18 +509,18 @@ async def image_generations( **{ "workflow": ComfyUIWorkflow( **{ - "workflow": app.state.config.COMFYUI_WORKFLOW, - "nodes": app.state.config.COMFYUI_WORKFLOW_NODES, + "workflow": request.app.state.config.COMFYUI_WORKFLOW, + "nodes": request.app.state.config.COMFYUI_WORKFLOW_NODES, } ), **data, } ) res = await comfyui_generate_image( - app.state.config.MODEL, + request.app.state.config.IMAGE_GENERATION_MODEL, form_data, user.id, - app.state.config.COMFYUI_BASE_URL, + request.app.state.config.COMFYUI_BASE_URL, ) log.debug(f"res: {res}") @@ -551,7 +537,8 @@ async def image_generations( log.debug(f"images: {images}") return images elif ( - app.state.config.ENGINE == "automatic1111" or app.state.config.ENGINE == "" + request.app.state.config.IMAGE_GENERATION_ENGINE == "automatic1111" + or request.app.state.config.IMAGE_GENERATION_ENGINE == "" ): if form_data.model: set_image_model(form_data.model) @@ -563,25 +550,25 @@ async def image_generations( "height": height, } - if app.state.config.IMAGE_STEPS is not None: - data["steps"] = app.state.config.IMAGE_STEPS + if request.app.state.config.IMAGE_STEPS is not None: + data["steps"] = request.app.state.config.IMAGE_STEPS if form_data.negative_prompt is not None: data["negative_prompt"] = form_data.negative_prompt - if app.state.config.AUTOMATIC1111_CFG_SCALE: - data["cfg_scale"] = app.state.config.AUTOMATIC1111_CFG_SCALE + if request.app.state.config.AUTOMATIC1111_CFG_SCALE: + data["cfg_scale"] = request.app.state.config.AUTOMATIC1111_CFG_SCALE - if app.state.config.AUTOMATIC1111_SAMPLER: - data["sampler_name"] = app.state.config.AUTOMATIC1111_SAMPLER + if request.app.state.config.AUTOMATIC1111_SAMPLER: + data["sampler_name"] = request.app.state.config.AUTOMATIC1111_SAMPLER - if app.state.config.AUTOMATIC1111_SCHEDULER: - data["scheduler"] = app.state.config.AUTOMATIC1111_SCHEDULER + if request.app.state.config.AUTOMATIC1111_SCHEDULER: + data["scheduler"] = request.app.state.config.AUTOMATIC1111_SCHEDULER # Use asyncio.to_thread for the requests.post call r = await asyncio.to_thread( requests.post, - url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/txt2img", + url=f"{request.app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/txt2img", json=data, headers={"authorization": get_automatic1111_api_auth()}, ) diff --git a/backend/open_webui/apps/webui/routers/knowledge.py b/backend/open_webui/routers/knowledge.py similarity index 96% rename from backend/open_webui/apps/webui/routers/knowledge.py rename to backend/open_webui/routers/knowledge.py index 21361c7f7..0dff2bc02 100644 --- a/backend/open_webui/apps/webui/routers/knowledge.py +++ b/backend/open_webui/routers/knowledge.py @@ -3,15 +3,15 @@ from pydantic import BaseModel from fastapi import APIRouter, Depends, HTTPException, status, Request import logging -from open_webui.apps.webui.models.knowledge import ( +from open_webui.models.knowledge import ( Knowledges, KnowledgeForm, KnowledgeResponse, KnowledgeUserResponse, ) -from open_webui.apps.webui.models.files import Files, FileModel -from open_webui.apps.retrieval.vector.connector import VECTOR_DB_CLIENT -from open_webui.apps.retrieval.main import BatchProcessFilesForm, process_file, ProcessFileForm, process_files_batch +from open_webui.models.files import Files, FileModel +from open_webui.retrieval.vector.connector import VECTOR_DB_CLIENT +from open_webui.routers.retrieval import process_file, ProcessFileForm, process_files_batch, BatchProcessFilesForm from open_webui.constants import ERROR_MESSAGES @@ -241,6 +241,7 @@ class KnowledgeFileIdForm(BaseModel): @router.post("/{id}/file/add", response_model=Optional[KnowledgeFilesResponse]) def add_file_to_knowledge_by_id( + request: Request, id: str, form_data: KnowledgeFileIdForm, user=Depends(get_verified_user), @@ -273,7 +274,9 @@ def add_file_to_knowledge_by_id( # Add content to the vector database try: - process_file(ProcessFileForm(file_id=form_data.file_id, collection_name=id)) + process_file( + request, ProcessFileForm(file_id=form_data.file_id, collection_name=id) + ) except Exception as e: log.debug(e) raise HTTPException( @@ -317,6 +320,7 @@ def add_file_to_knowledge_by_id( @router.post("/{id}/file/update", response_model=Optional[KnowledgeFilesResponse]) def update_file_from_knowledge_by_id( + request: Request, id: str, form_data: KnowledgeFileIdForm, user=Depends(get_verified_user), @@ -348,7 +352,9 @@ def update_file_from_knowledge_by_id( # Add content to the vector database try: - process_file(ProcessFileForm(file_id=form_data.file_id, collection_name=id)) + process_file( + request, ProcessFileForm(file_id=form_data.file_id, collection_name=id) + ) except Exception as e: raise HTTPException( status_code=status.HTTP_400_BAD_REQUEST, diff --git a/backend/open_webui/apps/webui/routers/memories.py b/backend/open_webui/routers/memories.py similarity index 97% rename from backend/open_webui/apps/webui/routers/memories.py rename to backend/open_webui/routers/memories.py index 60993607f..e72cf1445 100644 --- a/backend/open_webui/apps/webui/routers/memories.py +++ b/backend/open_webui/routers/memories.py @@ -3,8 +3,8 @@ from pydantic import BaseModel import logging from typing import Optional -from open_webui.apps.webui.models.memories import Memories, MemoryModel -from open_webui.apps.retrieval.vector.connector import VECTOR_DB_CLIENT +from open_webui.models.memories import Memories, MemoryModel +from open_webui.retrieval.vector.connector import VECTOR_DB_CLIENT from open_webui.utils.auth import get_verified_user from open_webui.env import SRC_LOG_LEVELS diff --git a/backend/open_webui/apps/webui/routers/models.py b/backend/open_webui/routers/models.py similarity index 99% rename from backend/open_webui/apps/webui/routers/models.py rename to backend/open_webui/routers/models.py index 2e073219a..db981a913 100644 --- a/backend/open_webui/apps/webui/routers/models.py +++ b/backend/open_webui/routers/models.py @@ -1,6 +1,6 @@ from typing import Optional -from open_webui.apps.webui.models.models import ( +from open_webui.models.models import ( ModelForm, ModelModel, ModelResponse, diff --git a/backend/open_webui/apps/ollama/main.py b/backend/open_webui/routers/ollama.py similarity index 63% rename from backend/open_webui/apps/ollama/main.py rename to backend/open_webui/routers/ollama.py index 48142fd9f..233e30ce5 100644 --- a/backend/open_webui/apps/ollama/main.py +++ b/backend/open_webui/routers/ollama.py @@ -1,3 +1,7 @@ +# TODO: Implement a more intelligent load balancing mechanism for distributing requests among multiple backend instances. +# Current implementation uses a simple round-robin approach (random.choice). Consider incorporating algorithms like weighted round-robin, +# least connections, or least response time for better resource utilization and performance optimization. + import asyncio import json import logging @@ -12,31 +16,23 @@ import aiohttp from aiocache import cached import requests -from open_webui.apps.webui.models.models import Models -from open_webui.config import ( - CORS_ALLOW_ORIGIN, - ENABLE_OLLAMA_API, - OLLAMA_BASE_URLS, - OLLAMA_API_CONFIGS, - UPLOAD_DIR, - AppConfig, -) -from open_webui.env import ( - AIOHTTP_CLIENT_TIMEOUT, - AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST, - BYPASS_MODEL_ACCESS_CONTROL, -) - -from open_webui.constants import ERROR_MESSAGES -from open_webui.env import ENV, SRC_LOG_LEVELS -from fastapi import Depends, FastAPI, File, HTTPException, Request, UploadFile +from fastapi import ( + Depends, + FastAPI, + File, + HTTPException, + Request, + UploadFile, + APIRouter, +) from fastapi.middleware.cors import CORSMiddleware from fastapi.responses import StreamingResponse from pydantic import BaseModel, ConfigDict from starlette.background import BackgroundTask +from open_webui.models.models import Models from open_webui.utils.misc import ( calculate_sha256, ) @@ -48,128 +44,37 @@ from open_webui.utils.payload import ( from open_webui.utils.auth import get_admin_user, get_verified_user from open_webui.utils.access_control import has_access + +from open_webui.config import ( + UPLOAD_DIR, +) +from open_webui.env import ( + ENV, + SRC_LOG_LEVELS, + AIOHTTP_CLIENT_TIMEOUT, + AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST, + BYPASS_MODEL_ACCESS_CONTROL, +) +from open_webui.constants import ERROR_MESSAGES + log = logging.getLogger(__name__) log.setLevel(SRC_LOG_LEVELS["OLLAMA"]) -app = FastAPI( - docs_url="/docs" if ENV == "dev" else None, - openapi_url="/openapi.json" if ENV == "dev" else None, - redoc_url=None, -) - -app.add_middleware( - CORSMiddleware, - allow_origins=CORS_ALLOW_ORIGIN, - allow_credentials=True, - allow_methods=["*"], - allow_headers=["*"], -) - -app.state.config = AppConfig() - -app.state.config.ENABLE_OLLAMA_API = ENABLE_OLLAMA_API -app.state.config.OLLAMA_BASE_URLS = OLLAMA_BASE_URLS -app.state.config.OLLAMA_API_CONFIGS = OLLAMA_API_CONFIGS +########################################## +# +# Utility functions +# +########################################## -# TODO: Implement a more intelligent load balancing mechanism for distributing requests among multiple backend instances. -# Current implementation uses a simple round-robin approach (random.choice). Consider incorporating algorithms like weighted round-robin, -# least connections, or least response time for better resource utilization and performance optimization. - - -@app.head("/") -@app.get("/") -async def get_status(): - return {"status": True} - - -class ConnectionVerificationForm(BaseModel): - url: str - key: Optional[str] = None - - -@app.post("/verify") -async def verify_connection( - form_data: ConnectionVerificationForm, user=Depends(get_admin_user) -): - url = form_data.url - key = form_data.key - - headers = {} - if key: - headers["Authorization"] = f"Bearer {key}" - - timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST) - async with aiohttp.ClientSession(timeout=timeout) as session: - try: - async with session.get(f"{url}/api/version", headers=headers) as r: - if r.status != 200: - # Extract response error details if available - error_detail = f"HTTP Error: {r.status}" - res = await r.json() - if "error" in res: - error_detail = f"External Error: {res['error']}" - raise Exception(error_detail) - - response_data = await r.json() - return response_data - - except aiohttp.ClientError as e: - # ClientError covers all aiohttp requests issues - log.exception(f"Client error: {str(e)}") - # Handle aiohttp-specific connection issues, timeout etc. - raise HTTPException( - status_code=500, detail="Open WebUI: Server Connection Error" - ) - except Exception as e: - log.exception(f"Unexpected error: {e}") - # Generic error handler in case parsing JSON or other steps fail - error_detail = f"Unexpected error: {str(e)}" - raise HTTPException(status_code=500, detail=error_detail) - - -@app.get("/config") -async def get_config(user=Depends(get_admin_user)): - return { - "ENABLE_OLLAMA_API": app.state.config.ENABLE_OLLAMA_API, - "OLLAMA_BASE_URLS": app.state.config.OLLAMA_BASE_URLS, - "OLLAMA_API_CONFIGS": app.state.config.OLLAMA_API_CONFIGS, - } - - -class OllamaConfigForm(BaseModel): - ENABLE_OLLAMA_API: Optional[bool] = None - OLLAMA_BASE_URLS: list[str] - OLLAMA_API_CONFIGS: dict - - -@app.post("/config/update") -async def update_config(form_data: OllamaConfigForm, user=Depends(get_admin_user)): - app.state.config.ENABLE_OLLAMA_API = form_data.ENABLE_OLLAMA_API - app.state.config.OLLAMA_BASE_URLS = form_data.OLLAMA_BASE_URLS - - app.state.config.OLLAMA_API_CONFIGS = form_data.OLLAMA_API_CONFIGS - - # Remove any extra configs - config_urls = app.state.config.OLLAMA_API_CONFIGS.keys() - for url in list(app.state.config.OLLAMA_BASE_URLS): - if url not in config_urls: - app.state.config.OLLAMA_API_CONFIGS.pop(url, None) - - return { - "ENABLE_OLLAMA_API": app.state.config.ENABLE_OLLAMA_API, - "OLLAMA_BASE_URLS": app.state.config.OLLAMA_BASE_URLS, - "OLLAMA_API_CONFIGS": app.state.config.OLLAMA_API_CONFIGS, - } - - -async def aiohttp_get(url, key=None): +async def send_get_request(url, key=None): timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST) try: - headers = {"Authorization": f"Bearer {key}"} if key else {} async with aiohttp.ClientSession(timeout=timeout, trust_env=True) as session: - async with session.get(url, headers=headers) as response: + async with session.get( + url, headers={**({"Authorization": f"Bearer {key}"} if key else {})} + ) as response: return await response.json() except Exception as e: # Handle connection error here @@ -177,46 +82,44 @@ async def aiohttp_get(url, key=None): return None -async def cleanup_response( - response: Optional[aiohttp.ClientResponse], - session: Optional[aiohttp.ClientSession], +async def send_post_request( + url: str, + payload: Union[str, bytes], + stream: bool = True, + key: Optional[str] = None, + content_type: Optional[str] = None, ): - if response: - response.close() - if session: - await session.close() + async def cleanup_response( + response: Optional[aiohttp.ClientResponse], + session: Optional[aiohttp.ClientSession], + ): + if response: + response.close() + if session: + await session.close() - -async def post_streaming_url( - url: str, payload: Union[str, bytes], stream: bool = True, content_type=None -): r = None try: session = aiohttp.ClientSession( trust_env=True, timeout=aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT) ) - parsed_url = urlparse(url) - base_url = f"{parsed_url.scheme}://{parsed_url.netloc}" - - api_config = app.state.config.OLLAMA_API_CONFIGS.get(base_url, {}) - key = api_config.get("key", None) - - headers = {"Content-Type": "application/json"} - if key: - headers["Authorization"] = f"Bearer {key}" - r = await session.post( url, data=payload, - headers=headers, + headers={ + "Content-Type": "application/json", + **({"Authorization": f"Bearer {key}"} if key else {}), + }, ) r.raise_for_status() if stream: response_headers = dict(r.headers) + if content_type: response_headers["Content-Type"] = content_type + return StreamingResponse( r.content, status_code=r.status, @@ -231,61 +134,146 @@ async def post_streaming_url( return res except Exception as e: - error_detail = "Open WebUI: Server Connection Error" + detail = None + if r is not None: try: res = await r.json() if "error" in res: - error_detail = f"Ollama: {res['error']}" + detail = f"Ollama: {res.get('error', 'Unknown error')}" except Exception: - error_detail = f"Ollama: {e}" + detail = f"Ollama: {e}" raise HTTPException( status_code=r.status if r else 500, - detail=error_detail, + detail=detail if detail else "Open WebUI: Server Connection Error", ) -def merge_models_lists(model_lists): - merged_models = {} +def get_api_key(url, configs): + parsed_url = urlparse(url) + base_url = f"{parsed_url.scheme}://{parsed_url.netloc}" + return configs.get(base_url, {}).get("key", None) - for idx, model_list in enumerate(model_lists): - if model_list is not None: - for model in model_list: - id = model["model"] - if id not in merged_models: - model["urls"] = [idx] - merged_models[id] = model - else: - merged_models[id]["urls"].append(idx) - return list(merged_models.values()) +########################################## +# +# API routes +# +########################################## + +router = APIRouter() + + +@router.head("/") +@router.get("/") +async def get_status(): + return {"status": True} + + +class ConnectionVerificationForm(BaseModel): + url: str + key: Optional[str] = None + + +@router.post("/verify") +async def verify_connection( + form_data: ConnectionVerificationForm, user=Depends(get_admin_user) +): + url = form_data.url + key = form_data.key + + async with aiohttp.ClientSession( + timeout=aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST) + ) as session: + try: + async with session.get( + f"{url}/api/version", + headers={**({"Authorization": f"Bearer {key}"} if key else {})}, + ) as r: + if r.status != 200: + detail = f"HTTP Error: {r.status}" + res = await r.json() + + if "error" in res: + detail = f"External Error: {res['error']}" + raise Exception(detail) + + data = await r.json() + return data + except aiohttp.ClientError as e: + log.exception(f"Client error: {str(e)}") + raise HTTPException( + status_code=500, detail="Open WebUI: Server Connection Error" + ) + except Exception as e: + log.exception(f"Unexpected error: {e}") + error_detail = f"Unexpected error: {str(e)}" + raise HTTPException(status_code=500, detail=error_detail) + + +@router.get("/config") +async def get_config(request: Request, user=Depends(get_admin_user)): + return { + "ENABLE_OLLAMA_API": request.app.state.config.ENABLE_OLLAMA_API, + "OLLAMA_BASE_URLS": request.app.state.config.OLLAMA_BASE_URLS, + "OLLAMA_API_CONFIGS": request.app.state.config.OLLAMA_API_CONFIGS, + } + + +class OllamaConfigForm(BaseModel): + ENABLE_OLLAMA_API: Optional[bool] = None + OLLAMA_BASE_URLS: list[str] + OLLAMA_API_CONFIGS: dict + + +@router.post("/config/update") +async def update_config( + request: Request, form_data: OllamaConfigForm, user=Depends(get_admin_user) +): + request.app.state.config.ENABLE_OLLAMA_API = form_data.ENABLE_OLLAMA_API + + request.app.state.config.OLLAMA_BASE_URLS = form_data.OLLAMA_BASE_URLS + request.app.state.config.OLLAMA_API_CONFIGS = form_data.OLLAMA_API_CONFIGS + + # Remove any extra configs + config_urls = request.app.state.config.OLLAMA_API_CONFIGS.keys() + for url in list(request.app.state.config.OLLAMA_BASE_URLS): + if url not in config_urls: + request.app.state.config.OLLAMA_API_CONFIGS.pop(url, None) + + return { + "ENABLE_OLLAMA_API": request.app.state.config.ENABLE_OLLAMA_API, + "OLLAMA_BASE_URLS": request.app.state.config.OLLAMA_BASE_URLS, + "OLLAMA_API_CONFIGS": request.app.state.config.OLLAMA_API_CONFIGS, + } @cached(ttl=3) -async def get_all_models(): +async def get_all_models(request: Request): log.info("get_all_models()") - if app.state.config.ENABLE_OLLAMA_API: - tasks = [] - for idx, url in enumerate(app.state.config.OLLAMA_BASE_URLS): - if url not in app.state.config.OLLAMA_API_CONFIGS: - tasks.append(aiohttp_get(f"{url}/api/tags")) + if request.app.state.config.ENABLE_OLLAMA_API: + request_tasks = [] + + for idx, url in enumerate(request.app.state.config.OLLAMA_BASE_URLS): + if url not in request.app.state.config.OLLAMA_API_CONFIGS: + request_tasks.append(send_get_request(f"{url}/api/tags")) else: - api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {}) + api_config = request.app.state.config.OLLAMA_API_CONFIGS.get(url, {}) enable = api_config.get("enable", True) key = api_config.get("key", None) if enable: - tasks.append(aiohttp_get(f"{url}/api/tags", key)) + request_tasks.append(send_get_request(f"{url}/api/tags", key)) else: - tasks.append(asyncio.ensure_future(asyncio.sleep(0, None))) + request_tasks.append(asyncio.ensure_future(asyncio.sleep(0, None))) - responses = await asyncio.gather(*tasks) + responses = await asyncio.gather(*request_tasks) for idx, response in enumerate(responses): if response: - url = app.state.config.OLLAMA_BASE_URLS[idx] - api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {}) + url = request.app.state.config.OLLAMA_BASE_URLS[idx] + api_config = request.app.state.config.OLLAMA_API_CONFIGS.get(url, {}) prefix_id = api_config.get("prefix_id", None) model_ids = api_config.get("model_ids", []) @@ -302,6 +290,21 @@ async def get_all_models(): for model in response.get("models", []): model["model"] = f"{prefix_id}.{model['model']}" + def merge_models_lists(model_lists): + merged_models = {} + + for idx, model_list in enumerate(model_lists): + if model_list is not None: + for model in model_list: + id = model["model"] + if id not in merged_models: + model["urls"] = [idx] + merged_models[id] = model + else: + merged_models[id]["urls"].append(idx) + + return list(merged_models.values()) + models = { "models": merge_models_lists( map( @@ -314,81 +317,87 @@ async def get_all_models(): else: models = {"models": []} + request.app.state.OLLAMA_MODELS = { + model["model"]: model for model in models["models"] + } return models -@app.get("/api/tags") -@app.get("/api/tags/{url_idx}") +async def get_filtered_models(models, user): + # Filter models based on user access control + filtered_models = [] + for model in models.get("models", []): + model_info = Models.get_model_by_id(model["model"]) + if model_info: + if user.id == model_info.user_id or has_access( + user.id, type="read", access_control=model_info.access_control + ): + filtered_models.append(model) + return filtered_models + + +@router.get("/api/tags") +@router.get("/api/tags/{url_idx}") async def get_ollama_tags( - url_idx: Optional[int] = None, user=Depends(get_verified_user) + request: Request, url_idx: Optional[int] = None, user=Depends(get_verified_user) ): models = [] + if url_idx is None: - models = await get_all_models() + models = await get_all_models(request) else: - url = app.state.config.OLLAMA_BASE_URLS[url_idx] - - parsed_url = urlparse(url) - base_url = f"{parsed_url.scheme}://{parsed_url.netloc}" - - api_config = app.state.config.OLLAMA_API_CONFIGS.get(base_url, {}) - key = api_config.get("key", None) - - headers = {} - if key: - headers["Authorization"] = f"Bearer {key}" + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] + key = get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS) r = None try: - r = requests.request(method="GET", url=f"{url}/api/tags", headers=headers) + r = requests.request( + method="GET", + url=f"{url}/api/tags", + headers={**({"Authorization": f"Bearer {key}"} if key else {})}, + ) r.raise_for_status() models = r.json() except Exception as e: log.exception(e) - error_detail = "Open WebUI: Server Connection Error" + + detail = None if r is not None: try: res = r.json() if "error" in res: - error_detail = f"Ollama: {res['error']}" + detail = f"Ollama: {res['error']}" except Exception: - error_detail = f"Ollama: {e}" + detail = f"Ollama: {e}" raise HTTPException( status_code=r.status_code if r else 500, - detail=error_detail, + detail=detail if detail else "Open WebUI: Server Connection Error", ) if user.role == "user" and not BYPASS_MODEL_ACCESS_CONTROL: - # Filter models based on user access control - filtered_models = [] - for model in models.get("models", []): - model_info = Models.get_model_by_id(model["model"]) - if model_info: - if user.id == model_info.user_id or has_access( - user.id, type="read", access_control=model_info.access_control - ): - filtered_models.append(model) - models["models"] = filtered_models + models["models"] = get_filtered_models(models, user) return models -@app.get("/api/version") -@app.get("/api/version/{url_idx}") -async def get_ollama_versions(url_idx: Optional[int] = None): - if app.state.config.ENABLE_OLLAMA_API: +@router.get("/api/version") +@router.get("/api/version/{url_idx}") +async def get_ollama_versions(request: Request, url_idx: Optional[int] = None): + if request.app.state.config.ENABLE_OLLAMA_API: if url_idx is None: # returns lowest version - tasks = [ - aiohttp_get( + request_tasks = [ + send_get_request( f"{url}/api/version", - app.state.config.OLLAMA_API_CONFIGS.get(url, {}).get("key", None), + request.app.state.config.OLLAMA_API_CONFIGS.get(url, {}).get( + "key", None + ), ) - for url in app.state.config.OLLAMA_BASE_URLS + for url in request.app.state.config.OLLAMA_BASE_URLS ] - responses = await asyncio.gather(*tasks) + responses = await asyncio.gather(*request_tasks) responses = list(filter(lambda x: x is not None, responses)) if len(responses) > 0: @@ -406,7 +415,7 @@ async def get_ollama_versions(url_idx: Optional[int] = None): detail=ERROR_MESSAGES.OLLAMA_NOT_FOUND, ) else: - url = app.state.config.OLLAMA_BASE_URLS[url_idx] + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] r = None try: @@ -416,39 +425,42 @@ async def get_ollama_versions(url_idx: Optional[int] = None): return r.json() except Exception as e: log.exception(e) - error_detail = "Open WebUI: Server Connection Error" + + detail = None if r is not None: try: res = r.json() if "error" in res: - error_detail = f"Ollama: {res['error']}" + detail = f"Ollama: {res['error']}" except Exception: - error_detail = f"Ollama: {e}" + detail = f"Ollama: {e}" raise HTTPException( status_code=r.status_code if r else 500, - detail=error_detail, + detail=detail if detail else "Open WebUI: Server Connection Error", ) else: return {"version": False} -@app.get("/api/ps") -async def get_ollama_loaded_models(user=Depends(get_verified_user)): +@router.get("/api/ps") +async def get_ollama_loaded_models(request: Request, user=Depends(get_verified_user)): """ List models that are currently loaded into Ollama memory, and which node they are loaded on. """ - if app.state.config.ENABLE_OLLAMA_API: - tasks = [ - aiohttp_get( + if request.app.state.config.ENABLE_OLLAMA_API: + request_tasks = [ + send_get_request( f"{url}/api/ps", - app.state.config.OLLAMA_API_CONFIGS.get(url, {}).get("key", None), + request.app.state.config.OLLAMA_API_CONFIGS.get(url, {}).get( + "key", None + ), ) - for url in app.state.config.OLLAMA_BASE_URLS + for url in request.app.state.config.OLLAMA_BASE_URLS ] - responses = await asyncio.gather(*tasks) + responses = await asyncio.gather(*request_tasks) - return dict(zip(app.state.config.OLLAMA_BASE_URLS, responses)) + return dict(zip(request.app.state.config.OLLAMA_BASE_URLS, responses)) else: return {} @@ -457,18 +469,25 @@ class ModelNameForm(BaseModel): name: str -@app.post("/api/pull") -@app.post("/api/pull/{url_idx}") +@router.post("/api/pull") +@router.post("/api/pull/{url_idx}") async def pull_model( - form_data: ModelNameForm, url_idx: int = 0, user=Depends(get_admin_user) + request: Request, + form_data: ModelNameForm, + url_idx: int = 0, + user=Depends(get_admin_user), ): - url = app.state.config.OLLAMA_BASE_URLS[url_idx] + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] log.info(f"url: {url}") # Admin should be able to pull models from any source payload = {**form_data.model_dump(exclude_none=True), "insecure": True} - return await post_streaming_url(f"{url}/api/pull", json.dumps(payload)) + return await send_post_request( + url=f"{url}/api/pull", + payload=json.dumps(payload), + key=get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS), + ) class PushModelForm(BaseModel): @@ -477,16 +496,17 @@ class PushModelForm(BaseModel): stream: Optional[bool] = None -@app.delete("/api/push") -@app.delete("/api/push/{url_idx}") +@router.delete("/api/push") +@router.delete("/api/push/{url_idx}") async def push_model( + request: Request, form_data: PushModelForm, url_idx: Optional[int] = None, user=Depends(get_admin_user), ): if url_idx is None: - model_list = await get_all_models() - models = {model["model"]: model for model in model_list["models"]} + await get_all_models(request) + models = request.app.state.OLLAMA_MODELS if form_data.name in models: url_idx = models[form_data.name]["urls"][0] @@ -496,11 +516,13 @@ async def push_model( detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.name), ) - url = app.state.config.OLLAMA_BASE_URLS[url_idx] + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] log.debug(f"url: {url}") - return await post_streaming_url( - f"{url}/api/push", form_data.model_dump_json(exclude_none=True).encode() + return await send_post_request( + url=f"{url}/api/push", + payload=form_data.model_dump_json(exclude_none=True).encode(), + key=get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS), ) @@ -511,17 +533,21 @@ class CreateModelForm(BaseModel): path: Optional[str] = None -@app.post("/api/create") -@app.post("/api/create/{url_idx}") +@router.post("/api/create") +@router.post("/api/create/{url_idx}") async def create_model( - form_data: CreateModelForm, url_idx: int = 0, user=Depends(get_admin_user) + request: Request, + form_data: CreateModelForm, + url_idx: int = 0, + user=Depends(get_admin_user), ): log.debug(f"form_data: {form_data}") - url = app.state.config.OLLAMA_BASE_URLS[url_idx] - log.info(f"url: {url}") + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] - return await post_streaming_url( - f"{url}/api/create", form_data.model_dump_json(exclude_none=True).encode() + return await send_post_request( + url=f"{url}/api/create", + payload=form_data.model_dump_json(exclude_none=True).encode(), + key=get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS), ) @@ -530,16 +556,17 @@ class CopyModelForm(BaseModel): destination: str -@app.post("/api/copy") -@app.post("/api/copy/{url_idx}") +@router.post("/api/copy") +@router.post("/api/copy/{url_idx}") async def copy_model( + request: Request, form_data: CopyModelForm, url_idx: Optional[int] = None, user=Depends(get_admin_user), ): if url_idx is None: - model_list = await get_all_models() - models = {model["model"]: model for model in model_list["models"]} + await get_all_models(request) + models = request.app.state.OLLAMA_MODELS if form_data.source in models: url_idx = models[form_data.source]["urls"][0] @@ -549,59 +576,52 @@ async def copy_model( detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.source), ) - url = app.state.config.OLLAMA_BASE_URLS[url_idx] - log.info(f"url: {url}") - - parsed_url = urlparse(url) - base_url = f"{parsed_url.scheme}://{parsed_url.netloc}" - - api_config = app.state.config.OLLAMA_API_CONFIGS.get(base_url, {}) - key = api_config.get("key", None) - - headers = {"Content-Type": "application/json"} - if key: - headers["Authorization"] = f"Bearer {key}" - - r = requests.request( - method="POST", - url=f"{url}/api/copy", - headers=headers, - data=form_data.model_dump_json(exclude_none=True).encode(), - ) + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] + key = get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS) try: + r = requests.request( + method="POST", + url=f"{url}/api/copy", + headers={ + "Content-Type": "application/json", + **({"Authorization": f"Bearer {key}"} if key else {}), + }, + data=form_data.model_dump_json(exclude_none=True).encode(), + ) r.raise_for_status() log.debug(f"r.text: {r.text}") - return True except Exception as e: log.exception(e) - error_detail = "Open WebUI: Server Connection Error" + + detail = None if r is not None: try: res = r.json() if "error" in res: - error_detail = f"Ollama: {res['error']}" + detail = f"Ollama: {res['error']}" except Exception: - error_detail = f"Ollama: {e}" + detail = f"Ollama: {e}" raise HTTPException( status_code=r.status_code if r else 500, - detail=error_detail, + detail=detail if detail else "Open WebUI: Server Connection Error", ) -@app.delete("/api/delete") -@app.delete("/api/delete/{url_idx}") +@router.delete("/api/delete") +@router.delete("/api/delete/{url_idx}") async def delete_model( + request: Request, form_data: ModelNameForm, url_idx: Optional[int] = None, user=Depends(get_admin_user), ): if url_idx is None: - model_list = await get_all_models() - models = {model["model"]: model for model in model_list["models"]} + await get_all_models(request) + models = request.app.state.OLLAMA_MODELS if form_data.name in models: url_idx = models[form_data.name]["urls"][0] @@ -611,52 +631,47 @@ async def delete_model( detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.name), ) - url = app.state.config.OLLAMA_BASE_URLS[url_idx] - log.info(f"url: {url}") + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] + key = get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS) - parsed_url = urlparse(url) - base_url = f"{parsed_url.scheme}://{parsed_url.netloc}" - - api_config = app.state.config.OLLAMA_API_CONFIGS.get(base_url, {}) - key = api_config.get("key", None) - - headers = {"Content-Type": "application/json"} - if key: - headers["Authorization"] = f"Bearer {key}" - - r = requests.request( - method="DELETE", - url=f"{url}/api/delete", - data=form_data.model_dump_json(exclude_none=True).encode(), - headers=headers, - ) try: + r = requests.request( + method="DELETE", + url=f"{url}/api/delete", + data=form_data.model_dump_json(exclude_none=True).encode(), + headers={ + "Content-Type": "application/json", + **({"Authorization": f"Bearer {key}"} if key else {}), + }, + ) r.raise_for_status() log.debug(f"r.text: {r.text}") - return True except Exception as e: log.exception(e) - error_detail = "Open WebUI: Server Connection Error" + + detail = None if r is not None: try: res = r.json() if "error" in res: - error_detail = f"Ollama: {res['error']}" + detail = f"Ollama: {res['error']}" except Exception: - error_detail = f"Ollama: {e}" + detail = f"Ollama: {e}" raise HTTPException( status_code=r.status_code if r else 500, - detail=error_detail, + detail=detail if detail else "Open WebUI: Server Connection Error", ) -@app.post("/api/show") -async def show_model_info(form_data: ModelNameForm, user=Depends(get_verified_user)): - model_list = await get_all_models() - models = {model["model"]: model for model in model_list["models"]} +@router.post("/api/show") +async def show_model_info( + request: Request, form_data: ModelNameForm, user=Depends(get_verified_user) +): + await get_all_models(request) + models = request.app.state.OLLAMA_MODELS if form_data.name not in models: raise HTTPException( @@ -665,53 +680,41 @@ async def show_model_info(form_data: ModelNameForm, user=Depends(get_verified_us ) url_idx = random.choice(models[form_data.name]["urls"]) - url = app.state.config.OLLAMA_BASE_URLS[url_idx] - log.info(f"url: {url}") - parsed_url = urlparse(url) - base_url = f"{parsed_url.scheme}://{parsed_url.netloc}" + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] + key = get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS) - api_config = app.state.config.OLLAMA_API_CONFIGS.get(base_url, {}) - key = api_config.get("key", None) - - headers = {"Content-Type": "application/json"} - if key: - headers["Authorization"] = f"Bearer {key}" - - r = requests.request( - method="POST", - url=f"{url}/api/show", - headers=headers, - data=form_data.model_dump_json(exclude_none=True).encode(), - ) try: + r = requests.request( + method="POST", + url=f"{url}/api/show", + headers={ + "Content-Type": "application/json", + **({"Authorization": f"Bearer {key}"} if key else {}), + }, + data=form_data.model_dump_json(exclude_none=True).encode(), + ) r.raise_for_status() return r.json() except Exception as e: log.exception(e) - error_detail = "Open WebUI: Server Connection Error" + + detail = None if r is not None: try: res = r.json() if "error" in res: - error_detail = f"Ollama: {res['error']}" + detail = f"Ollama: {res['error']}" except Exception: - error_detail = f"Ollama: {e}" + detail = f"Ollama: {e}" raise HTTPException( status_code=r.status_code if r else 500, - detail=error_detail, + detail=detail if detail else "Open WebUI: Server Connection Error", ) -class GenerateEmbeddingsForm(BaseModel): - model: str - prompt: str - options: Optional[dict] = None - keep_alive: Optional[Union[int, str]] = None - - class GenerateEmbedForm(BaseModel): model: str input: list[str] | str @@ -720,105 +723,19 @@ class GenerateEmbedForm(BaseModel): keep_alive: Optional[Union[int, str]] = None -@app.post("/api/embed") -@app.post("/api/embed/{url_idx}") -async def generate_embeddings( +@router.post("/api/embed") +@router.post("/api/embed/{url_idx}") +async def embed( + request: Request, form_data: GenerateEmbedForm, url_idx: Optional[int] = None, user=Depends(get_verified_user), -): - return await generate_ollama_batch_embeddings(form_data, url_idx) - - -@app.post("/api/embeddings") -@app.post("/api/embeddings/{url_idx}") -async def generate_embeddings( - form_data: GenerateEmbeddingsForm, - url_idx: Optional[int] = None, - user=Depends(get_verified_user), -): - return await generate_ollama_embeddings(form_data=form_data, url_idx=url_idx) - - -async def generate_ollama_embeddings( - form_data: GenerateEmbeddingsForm, - url_idx: Optional[int] = None, -): - log.info(f"generate_ollama_embeddings {form_data}") - - if url_idx is None: - model_list = await get_all_models() - models = {model["model"]: model for model in model_list["models"]} - - model = form_data.model - - if ":" not in model: - model = f"{model}:latest" - - if model in models: - url_idx = random.choice(models[model]["urls"]) - else: - raise HTTPException( - status_code=400, - detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model), - ) - - url = app.state.config.OLLAMA_BASE_URLS[url_idx] - log.info(f"url: {url}") - - parsed_url = urlparse(url) - base_url = f"{parsed_url.scheme}://{parsed_url.netloc}" - - api_config = app.state.config.OLLAMA_API_CONFIGS.get(base_url, {}) - key = api_config.get("key", None) - - headers = {"Content-Type": "application/json"} - if key: - headers["Authorization"] = f"Bearer {key}" - - r = requests.request( - method="POST", - url=f"{url}/api/embeddings", - headers=headers, - data=form_data.model_dump_json(exclude_none=True).encode(), - ) - try: - r.raise_for_status() - - data = r.json() - - log.info(f"generate_ollama_embeddings {data}") - - if "embedding" in data: - return data - else: - raise Exception("Something went wrong :/") - except Exception as e: - log.exception(e) - error_detail = "Open WebUI: Server Connection Error" - if r is not None: - try: - res = r.json() - if "error" in res: - error_detail = f"Ollama: {res['error']}" - except Exception: - error_detail = f"Ollama: {e}" - - raise HTTPException( - status_code=r.status_code if r else 500, - detail=error_detail, - ) - - -async def generate_ollama_batch_embeddings( - form_data: GenerateEmbedForm, - url_idx: Optional[int] = None, ): log.info(f"generate_ollama_batch_embeddings {form_data}") if url_idx is None: - model_list = await get_all_models() - models = {model["model"]: model for model in model_list["models"]} + await get_all_models(request) + models = request.app.state.OLLAMA_MODELS model = form_data.model @@ -833,48 +750,108 @@ async def generate_ollama_batch_embeddings( detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model), ) - url = app.state.config.OLLAMA_BASE_URLS[url_idx] - log.info(f"url: {url}") + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] + key = get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS) - parsed_url = urlparse(url) - base_url = f"{parsed_url.scheme}://{parsed_url.netloc}" - - api_config = app.state.config.OLLAMA_API_CONFIGS.get(base_url, {}) - key = api_config.get("key", None) - - headers = {"Content-Type": "application/json"} - if key: - headers["Authorization"] = f"Bearer {key}" - - r = requests.request( - method="POST", - url=f"{url}/api/embed", - headers=headers, - data=form_data.model_dump_json(exclude_none=True).encode(), - ) try: + r = requests.request( + method="POST", + url=f"{url}/api/embed", + headers={ + "Content-Type": "application/json", + **({"Authorization": f"Bearer {key}"} if key else {}), + }, + data=form_data.model_dump_json(exclude_none=True).encode(), + ) r.raise_for_status() data = r.json() - - log.info(f"generate_ollama_batch_embeddings {data}") - - if "embeddings" in data: - return data - else: - raise Exception("Something went wrong :/") + return data except Exception as e: log.exception(e) - error_detail = "Open WebUI: Server Connection Error" + + detail = None if r is not None: try: res = r.json() if "error" in res: - error_detail = f"Ollama: {res['error']}" + detail = f"Ollama: {res['error']}" except Exception: - error_detail = f"Ollama: {e}" + detail = f"Ollama: {e}" - raise Exception(error_detail) + raise HTTPException( + status_code=r.status_code if r else 500, + detail=detail if detail else "Open WebUI: Server Connection Error", + ) + + +class GenerateEmbeddingsForm(BaseModel): + model: str + prompt: str + options: Optional[dict] = None + keep_alive: Optional[Union[int, str]] = None + + +@router.post("/api/embeddings") +@router.post("/api/embeddings/{url_idx}") +async def embeddings( + request: Request, + form_data: GenerateEmbeddingsForm, + url_idx: Optional[int] = None, + user=Depends(get_verified_user), +): + log.info(f"generate_ollama_embeddings {form_data}") + + if url_idx is None: + await get_all_models(request) + models = request.app.state.OLLAMA_MODELS + + model = form_data.model + + if ":" not in model: + model = f"{model}:latest" + + if model in models: + url_idx = random.choice(models[model]["urls"]) + else: + raise HTTPException( + status_code=400, + detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model), + ) + + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] + key = get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS) + + try: + r = requests.request( + method="POST", + url=f"{url}/api/embeddings", + headers={ + "Content-Type": "application/json", + **({"Authorization": f"Bearer {key}"} if key else {}), + }, + data=form_data.model_dump_json(exclude_none=True).encode(), + ) + r.raise_for_status() + + data = r.json() + return data + except Exception as e: + log.exception(e) + + detail = None + if r is not None: + try: + res = r.json() + if "error" in res: + detail = f"Ollama: {res['error']}" + except Exception: + detail = f"Ollama: {e}" + + raise HTTPException( + status_code=r.status_code if r else 500, + detail=detail if detail else "Open WebUI: Server Connection Error", + ) class GenerateCompletionForm(BaseModel): @@ -892,16 +869,17 @@ class GenerateCompletionForm(BaseModel): keep_alive: Optional[Union[int, str]] = None -@app.post("/api/generate") -@app.post("/api/generate/{url_idx}") +@router.post("/api/generate") +@router.post("/api/generate/{url_idx}") async def generate_completion( + request: Request, form_data: GenerateCompletionForm, url_idx: Optional[int] = None, user=Depends(get_verified_user), ): if url_idx is None: - model_list = await get_all_models() - models = {model["model"]: model for model in model_list["models"]} + await get_all_models(request) + models = request.app.state.OLLAMA_MODELS model = form_data.model @@ -916,15 +894,17 @@ async def generate_completion( detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model), ) - url = app.state.config.OLLAMA_BASE_URLS[url_idx] - api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {}) + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] + api_config = request.app.state.config.OLLAMA_API_CONFIGS.get(url, {}) + prefix_id = api_config.get("prefix_id", None) if prefix_id: form_data.model = form_data.model.replace(f"{prefix_id}.", "") - log.info(f"url: {url}") - return await post_streaming_url( - f"{url}/api/generate", form_data.model_dump_json(exclude_none=True).encode() + return await send_post_request( + url=f"{url}/api/generate", + payload=form_data.model_dump_json(exclude_none=True).encode(), + key=get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS), ) @@ -944,25 +924,24 @@ class GenerateChatCompletionForm(BaseModel): keep_alive: Optional[Union[int, str]] = None -async def get_ollama_url(url_idx: Optional[int], model: str): +async def get_ollama_url(request: Request, model: str, url_idx: Optional[int] = None): if url_idx is None: - model_list = await get_all_models() - models = {model["model"]: model for model in model_list["models"]} - + models = request.app.state.OLLAMA_MODELS if model not in models: raise HTTPException( status_code=400, detail=ERROR_MESSAGES.MODEL_NOT_FOUND(model), ) - url_idx = random.choice(models[model]["urls"]) - url = app.state.config.OLLAMA_BASE_URLS[url_idx] + url_idx = random.choice(models[model].get("urls", [])) + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] return url -@app.post("/api/chat") -@app.post("/api/chat/{url_idx}") +@router.post("/api/chat") +@router.post("/api/chat/{url_idx}") async def generate_chat_completion( - form_data: GenerateChatCompletionForm, + request: Request, + form_data: dict, url_idx: Optional[int] = None, user=Depends(get_verified_user), bypass_filter: Optional[bool] = False, @@ -970,8 +949,16 @@ async def generate_chat_completion( if BYPASS_MODEL_ACCESS_CONTROL: bypass_filter = True + try: + form_data = GenerateChatCompletionForm(**form_data) + except Exception as e: + log.exception(e) + raise HTTPException( + status_code=400, + detail=str(e), + ) + payload = {**form_data.model_dump(exclude_none=True)} - log.debug(f"generate_chat_completion() - 1.payload = {payload}") if "metadata" in payload: del payload["metadata"] @@ -1015,22 +1002,18 @@ async def generate_chat_completion( if ":" not in payload["model"]: payload["model"] = f"{payload['model']}:latest" - url = await get_ollama_url(url_idx, payload["model"]) - log.info(f"url: {url}") - log.debug(f"generate_chat_completion() - 2.payload = {payload}") + url = await get_ollama_url(request, payload["model"], url_idx) + api_config = request.app.state.config.OLLAMA_API_CONFIGS.get(url, {}) - parsed_url = urlparse(url) - base_url = f"{parsed_url.scheme}://{parsed_url.netloc}" - - api_config = app.state.config.OLLAMA_API_CONFIGS.get(base_url, {}) prefix_id = api_config.get("prefix_id", None) if prefix_id: payload["model"] = payload["model"].replace(f"{prefix_id}.", "") - return await post_streaming_url( - f"{url}/api/chat", - json.dumps(payload), + return await send_post_request( + url=f"{url}/api/chat", + payload=json.dumps(payload), stream=form_data.stream, + key=get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS), content_type="application/x-ndjson", ) @@ -1062,10 +1045,13 @@ class OpenAICompletionForm(BaseModel): model_config = ConfigDict(extra="allow") -@app.post("/v1/completions") -@app.post("/v1/completions/{url_idx}") +@router.post("/v1/completions") +@router.post("/v1/completions/{url_idx}") async def generate_openai_completion( - form_data: dict, url_idx: Optional[int] = None, user=Depends(get_verified_user) + request: Request, + form_data: dict, + url_idx: Optional[int] = None, + user=Depends(get_verified_user), ): try: form_data = OpenAICompletionForm(**form_data) @@ -1115,25 +1101,26 @@ async def generate_openai_completion( if ":" not in payload["model"]: payload["model"] = f"{payload['model']}:latest" - url = await get_ollama_url(url_idx, payload["model"]) - log.info(f"url: {url}") + url = await get_ollama_url(request, payload["model"], url_idx) + api_config = request.app.state.config.OLLAMA_API_CONFIGS.get(url, {}) - api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {}) prefix_id = api_config.get("prefix_id", None) if prefix_id: payload["model"] = payload["model"].replace(f"{prefix_id}.", "") - return await post_streaming_url( - f"{url}/v1/completions", - json.dumps(payload), + return await send_post_request( + url=f"{url}/v1/completions", + payload=json.dumps(payload), stream=payload.get("stream", False), + key=get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS), ) -@app.post("/v1/chat/completions") -@app.post("/v1/chat/completions/{url_idx}") +@router.post("/v1/chat/completions") +@router.post("/v1/chat/completions/{url_idx}") async def generate_openai_chat_completion( + request: Request, form_data: dict, url_idx: Optional[int] = None, user=Depends(get_verified_user), @@ -1188,31 +1175,32 @@ async def generate_openai_chat_completion( if ":" not in payload["model"]: payload["model"] = f"{payload['model']}:latest" - url = await get_ollama_url(url_idx, payload["model"]) - log.info(f"url: {url}") + url = await get_ollama_url(request, payload["model"], url_idx) + api_config = request.app.state.config.OLLAMA_API_CONFIGS.get(url, {}) - api_config = app.state.config.OLLAMA_API_CONFIGS.get(url, {}) prefix_id = api_config.get("prefix_id", None) if prefix_id: payload["model"] = payload["model"].replace(f"{prefix_id}.", "") - return await post_streaming_url( - f"{url}/v1/chat/completions", - json.dumps(payload), + return await send_post_request( + url=f"{url}/v1/chat/completions", + payload=json.dumps(payload), stream=payload.get("stream", False), + key=get_api_key(url, request.app.state.config.OLLAMA_API_CONFIGS), ) -@app.get("/v1/models") -@app.get("/v1/models/{url_idx}") +@router.get("/v1/models") +@router.get("/v1/models/{url_idx}") async def get_openai_models( + request: Request, url_idx: Optional[int] = None, user=Depends(get_verified_user), ): models = [] if url_idx is None: - model_list = await get_all_models() + model_list = await get_all_models(request) models = [ { "id": model["model"], @@ -1224,7 +1212,7 @@ async def get_openai_models( ] else: - url = app.state.config.OLLAMA_BASE_URLS[url_idx] + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] try: r = requests.request(method="GET", url=f"{url}/api/tags") r.raise_for_status() @@ -1348,9 +1336,10 @@ async def download_file_stream( # url = "https://huggingface.co/TheBloke/stablelm-zephyr-3b-GGUF/resolve/main/stablelm-zephyr-3b.Q2_K.gguf" -@app.post("/models/download") -@app.post("/models/download/{url_idx}") +@router.post("/models/download") +@router.post("/models/download/{url_idx}") async def download_model( + request: Request, form_data: UrlForm, url_idx: Optional[int] = None, user=Depends(get_admin_user), @@ -1365,7 +1354,7 @@ async def download_model( if url_idx is None: url_idx = 0 - url = app.state.config.OLLAMA_BASE_URLS[url_idx] + url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] file_name = parse_huggingface_url(form_data.url) @@ -1379,16 +1368,17 @@ async def download_model( return None -@app.post("/models/upload") -@app.post("/models/upload/{url_idx}") +@router.post("/models/upload") +@router.post("/models/upload/{url_idx}") def upload_model( + request: Request, file: UploadFile = File(...), url_idx: Optional[int] = None, user=Depends(get_admin_user), ): if url_idx is None: url_idx = 0 - ollama_url = app.state.config.OLLAMA_BASE_URLS[url_idx] + ollama_url = request.app.state.config.OLLAMA_BASE_URLS[url_idx] file_path = f"{UPLOAD_DIR}/{file.filename}" diff --git a/backend/open_webui/apps/openai/main.py b/backend/open_webui/routers/openai.py similarity index 52% rename from backend/open_webui/apps/openai/main.py rename to backend/open_webui/routers/openai.py index b64e7b28d..f7f78be85 100644 --- a/backend/open_webui/apps/openai/main.py +++ b/backend/open_webui/routers/openai.py @@ -10,15 +10,15 @@ from aiocache import cached import requests -from open_webui.apps.webui.models.models import Models +from fastapi import Depends, FastAPI, HTTPException, Request, APIRouter +from fastapi.middleware.cors import CORSMiddleware +from fastapi.responses import FileResponse, StreamingResponse +from pydantic import BaseModel +from starlette.background import BackgroundTask + +from open_webui.models.models import Models from open_webui.config import ( CACHE_DIR, - CORS_ALLOW_ORIGIN, - ENABLE_OPENAI_API, - OPENAI_API_BASE_URLS, - OPENAI_API_KEYS, - OPENAI_API_CONFIGS, - AppConfig, ) from open_webui.env import ( AIOHTTP_CLIENT_TIMEOUT, @@ -29,11 +29,7 @@ from open_webui.env import ( from open_webui.constants import ERROR_MESSAGES from open_webui.env import ENV, SRC_LOG_LEVELS -from fastapi import Depends, FastAPI, HTTPException, Request -from fastapi.middleware.cors import CORSMiddleware -from fastapi.responses import FileResponse, StreamingResponse -from pydantic import BaseModel -from starlette.background import BackgroundTask + from open_webui.utils.payload import ( apply_model_params_to_body_openai, @@ -48,36 +44,69 @@ log = logging.getLogger(__name__) log.setLevel(SRC_LOG_LEVELS["OPENAI"]) -app = FastAPI( - docs_url="/docs" if ENV == "dev" else None, - openapi_url="/openapi.json" if ENV == "dev" else None, - redoc_url=None, -) +########################################## +# +# Utility functions +# +########################################## -app.add_middleware( - CORSMiddleware, - allow_origins=CORS_ALLOW_ORIGIN, - allow_credentials=True, - allow_methods=["*"], - allow_headers=["*"], -) - -app.state.config = AppConfig() - -app.state.config.ENABLE_OPENAI_API = ENABLE_OPENAI_API -app.state.config.OPENAI_API_BASE_URLS = OPENAI_API_BASE_URLS -app.state.config.OPENAI_API_KEYS = OPENAI_API_KEYS -app.state.config.OPENAI_API_CONFIGS = OPENAI_API_CONFIGS +async def send_get_request(url, key=None): + timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST) + try: + async with aiohttp.ClientSession(timeout=timeout, trust_env=True) as session: + async with session.get( + url, headers={**({"Authorization": f"Bearer {key}"} if key else {})} + ) as response: + return await response.json() + except Exception as e: + # Handle connection error here + log.error(f"Connection error: {e}") + return None -@app.get("/config") -async def get_config(user=Depends(get_admin_user)): +async def cleanup_response( + response: Optional[aiohttp.ClientResponse], + session: Optional[aiohttp.ClientSession], +): + if response: + response.close() + if session: + await session.close() + + +def openai_o1_handler(payload): + """ + Handle O1 specific parameters + """ + if "max_tokens" in payload: + # Remove "max_tokens" from the payload + payload["max_completion_tokens"] = payload["max_tokens"] + del payload["max_tokens"] + + # Fix: O1 does not support the "system" parameter, Modify "system" to "user" + if payload["messages"][0]["role"] == "system": + payload["messages"][0]["role"] = "user" + + return payload + + +########################################## +# +# API routes +# +########################################## + +router = APIRouter() + + +@router.get("/config") +async def get_config(request: Request, user=Depends(get_admin_user)): return { - "ENABLE_OPENAI_API": app.state.config.ENABLE_OPENAI_API, - "OPENAI_API_BASE_URLS": app.state.config.OPENAI_API_BASE_URLS, - "OPENAI_API_KEYS": app.state.config.OPENAI_API_KEYS, - "OPENAI_API_CONFIGS": app.state.config.OPENAI_API_CONFIGS, + "ENABLE_OPENAI_API": request.app.state.config.ENABLE_OPENAI_API, + "OPENAI_API_BASE_URLS": request.app.state.config.OPENAI_API_BASE_URLS, + "OPENAI_API_KEYS": request.app.state.config.OPENAI_API_KEYS, + "OPENAI_API_CONFIGS": request.app.state.config.OPENAI_API_CONFIGS, } @@ -88,50 +117,56 @@ class OpenAIConfigForm(BaseModel): OPENAI_API_CONFIGS: dict -@app.post("/config/update") -async def update_config(form_data: OpenAIConfigForm, user=Depends(get_admin_user)): - app.state.config.ENABLE_OPENAI_API = form_data.ENABLE_OPENAI_API - - app.state.config.OPENAI_API_BASE_URLS = form_data.OPENAI_API_BASE_URLS - app.state.config.OPENAI_API_KEYS = form_data.OPENAI_API_KEYS +@router.post("/config/update") +async def update_config( + request: Request, form_data: OpenAIConfigForm, user=Depends(get_admin_user) +): + request.app.state.config.ENABLE_OPENAI_API = form_data.ENABLE_OPENAI_API + request.app.state.config.OPENAI_API_BASE_URLS = form_data.OPENAI_API_BASE_URLS + request.app.state.config.OPENAI_API_KEYS = form_data.OPENAI_API_KEYS # Check if API KEYS length is same than API URLS length - if len(app.state.config.OPENAI_API_KEYS) != len( - app.state.config.OPENAI_API_BASE_URLS + if len(request.app.state.config.OPENAI_API_KEYS) != len( + request.app.state.config.OPENAI_API_BASE_URLS ): - if len(app.state.config.OPENAI_API_KEYS) > len( - app.state.config.OPENAI_API_BASE_URLS + if len(request.app.state.config.OPENAI_API_KEYS) > len( + request.app.state.config.OPENAI_API_BASE_URLS ): - app.state.config.OPENAI_API_KEYS = app.state.config.OPENAI_API_KEYS[ - : len(app.state.config.OPENAI_API_BASE_URLS) - ] + request.app.state.config.OPENAI_API_KEYS = ( + request.app.state.config.OPENAI_API_KEYS[ + : len(request.app.state.config.OPENAI_API_BASE_URLS) + ] + ) else: - app.state.config.OPENAI_API_KEYS += [""] * ( - len(app.state.config.OPENAI_API_BASE_URLS) - - len(app.state.config.OPENAI_API_KEYS) + request.app.state.config.OPENAI_API_KEYS += [""] * ( + len(request.app.state.config.OPENAI_API_BASE_URLS) + - len(request.app.state.config.OPENAI_API_KEYS) ) - app.state.config.OPENAI_API_CONFIGS = form_data.OPENAI_API_CONFIGS + request.app.state.config.OPENAI_API_CONFIGS = form_data.OPENAI_API_CONFIGS # Remove any extra configs - config_urls = app.state.config.OPENAI_API_CONFIGS.keys() - for idx, url in enumerate(app.state.config.OPENAI_API_BASE_URLS): + config_urls = request.app.state.config.OPENAI_API_CONFIGS.keys() + for idx, url in enumerate(request.app.state.config.OPENAI_API_BASE_URLS): if url not in config_urls: - app.state.config.OPENAI_API_CONFIGS.pop(url, None) + request.app.state.config.OPENAI_API_CONFIGS.pop(url, None) return { - "ENABLE_OPENAI_API": app.state.config.ENABLE_OPENAI_API, - "OPENAI_API_BASE_URLS": app.state.config.OPENAI_API_BASE_URLS, - "OPENAI_API_KEYS": app.state.config.OPENAI_API_KEYS, - "OPENAI_API_CONFIGS": app.state.config.OPENAI_API_CONFIGS, + "ENABLE_OPENAI_API": request.app.state.config.ENABLE_OPENAI_API, + "OPENAI_API_BASE_URLS": request.app.state.config.OPENAI_API_BASE_URLS, + "OPENAI_API_KEYS": request.app.state.config.OPENAI_API_KEYS, + "OPENAI_API_CONFIGS": request.app.state.config.OPENAI_API_CONFIGS, } -@app.post("/audio/speech") +@router.post("/audio/speech") async def speech(request: Request, user=Depends(get_verified_user)): idx = None try: - idx = app.state.config.OPENAI_API_BASE_URLS.index("https://api.openai.com/v1") + idx = request.app.state.config.OPENAI_API_BASE_URLS.index( + "https://api.openai.com/v1" + ) + body = await request.body() name = hashlib.sha256(body).hexdigest() @@ -144,23 +179,35 @@ async def speech(request: Request, user=Depends(get_verified_user)): if file_path.is_file(): return FileResponse(file_path) - headers = {} - headers["Authorization"] = f"Bearer {app.state.config.OPENAI_API_KEYS[idx]}" - headers["Content-Type"] = "application/json" - if "openrouter.ai" in app.state.config.OPENAI_API_BASE_URLS[idx]: - headers["HTTP-Referer"] = "https://openwebui.com/" - headers["X-Title"] = "Open WebUI" - if ENABLE_FORWARD_USER_INFO_HEADERS: - headers["X-OpenWebUI-User-Name"] = user.name - headers["X-OpenWebUI-User-Id"] = user.id - headers["X-OpenWebUI-User-Email"] = user.email - headers["X-OpenWebUI-User-Role"] = user.role + url = request.app.state.config.OPENAI_API_BASE_URLS[idx] + r = None try: r = requests.post( - url=f"{app.state.config.OPENAI_API_BASE_URLS[idx]}/audio/speech", + url=f"{url}/audio/speech", data=body, - headers=headers, + headers={ + "Content-Type": "application/json", + "Authorization": f"Bearer {request.app.state.config.OPENAI_API_KEYS[idx]}", + **( + { + "HTTP-Referer": "https://openwebui.com/", + "X-Title": "Open WebUI", + } + if "openrouter.ai" in url + else {} + ), + **( + { + "X-OpenWebUI-User-Name": user.name, + "X-OpenWebUI-User-Id": user.id, + "X-OpenWebUI-User-Email": user.email, + "X-OpenWebUI-User-Role": user.role, + } + if ENABLE_FORWARD_USER_INFO_HEADERS + else {} + ), + }, stream=True, ) @@ -179,115 +226,62 @@ async def speech(request: Request, user=Depends(get_verified_user)): except Exception as e: log.exception(e) - error_detail = "Open WebUI: Server Connection Error" + + detail = None if r is not None: try: res = r.json() if "error" in res: - error_detail = f"External: {res['error']}" + detail = f"External: {res['error']}" except Exception: - error_detail = f"External: {e}" + detail = f"External: {e}" raise HTTPException( - status_code=r.status_code if r else 500, detail=error_detail + status_code=r.status_code if r else 500, + detail=detail if detail else "Open WebUI: Server Connection Error", ) except ValueError: raise HTTPException(status_code=401, detail=ERROR_MESSAGES.OPENAI_NOT_FOUND) -async def aiohttp_get(url, key=None): - timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST) - try: - headers = {"Authorization": f"Bearer {key}"} if key else {} - async with aiohttp.ClientSession(timeout=timeout, trust_env=True) as session: - async with session.get(url, headers=headers) as response: - return await response.json() - except Exception as e: - # Handle connection error here - log.error(f"Connection error: {e}") - return None - - -async def cleanup_response( - response: Optional[aiohttp.ClientResponse], - session: Optional[aiohttp.ClientSession], -): - if response: - response.close() - if session: - await session.close() - - -def merge_models_lists(model_lists): - log.debug(f"merge_models_lists {model_lists}") - merged_list = [] - - for idx, models in enumerate(model_lists): - if models is not None and "error" not in models: - merged_list.extend( - [ - { - **model, - "name": model.get("name", model["id"]), - "owned_by": "openai", - "openai": model, - "urlIdx": idx, - } - for model in models - if "api.openai.com" - not in app.state.config.OPENAI_API_BASE_URLS[idx] - or not any( - name in model["id"] - for name in [ - "babbage", - "dall-e", - "davinci", - "embedding", - "tts", - "whisper", - ] - ) - ] - ) - - return merged_list - - -async def get_all_models_responses() -> list: - if not app.state.config.ENABLE_OPENAI_API: +async def get_all_models_responses(request: Request) -> list: + if not request.app.state.config.ENABLE_OPENAI_API: return [] # Check if API KEYS length is same than API URLS length - num_urls = len(app.state.config.OPENAI_API_BASE_URLS) - num_keys = len(app.state.config.OPENAI_API_KEYS) + num_urls = len(request.app.state.config.OPENAI_API_BASE_URLS) + num_keys = len(request.app.state.config.OPENAI_API_KEYS) if num_keys != num_urls: # if there are more keys than urls, remove the extra keys if num_keys > num_urls: - new_keys = app.state.config.OPENAI_API_KEYS[:num_urls] - app.state.config.OPENAI_API_KEYS = new_keys + new_keys = request.app.state.config.OPENAI_API_KEYS[:num_urls] + request.app.state.config.OPENAI_API_KEYS = new_keys # if there are more urls than keys, add empty keys else: - app.state.config.OPENAI_API_KEYS += [""] * (num_urls - num_keys) + request.app.state.config.OPENAI_API_KEYS += [""] * (num_urls - num_keys) - tasks = [] - for idx, url in enumerate(app.state.config.OPENAI_API_BASE_URLS): - if url not in app.state.config.OPENAI_API_CONFIGS: - tasks.append( - aiohttp_get(f"{url}/models", app.state.config.OPENAI_API_KEYS[idx]) + request_tasks = [] + for idx, url in enumerate(request.app.state.config.OPENAI_API_BASE_URLS): + if url not in request.app.state.config.OPENAI_API_CONFIGS: + request_tasks.append( + send_get_request( + f"{url}/models", request.app.state.config.OPENAI_API_KEYS[idx] + ) ) else: - api_config = app.state.config.OPENAI_API_CONFIGS.get(url, {}) + api_config = request.app.state.config.OPENAI_API_CONFIGS.get(url, {}) enable = api_config.get("enable", True) model_ids = api_config.get("model_ids", []) if enable: if len(model_ids) == 0: - tasks.append( - aiohttp_get( - f"{url}/models", app.state.config.OPENAI_API_KEYS[idx] + request_tasks.append( + send_get_request( + f"{url}/models", + request.app.state.config.OPENAI_API_KEYS[idx], ) ) else: @@ -305,16 +299,18 @@ async def get_all_models_responses() -> list: ], } - tasks.append(asyncio.ensure_future(asyncio.sleep(0, model_list))) + request_tasks.append( + asyncio.ensure_future(asyncio.sleep(0, model_list)) + ) else: - tasks.append(asyncio.ensure_future(asyncio.sleep(0, None))) + request_tasks.append(asyncio.ensure_future(asyncio.sleep(0, None))) - responses = await asyncio.gather(*tasks) + responses = await asyncio.gather(*request_tasks) for idx, response in enumerate(responses): if response: - url = app.state.config.OPENAI_API_BASE_URLS[idx] - api_config = app.state.config.OPENAI_API_CONFIGS.get(url, {}) + url = request.app.state.config.OPENAI_API_BASE_URLS[idx] + api_config = request.app.state.config.OPENAI_API_CONFIGS.get(url, {}) prefix_id = api_config.get("prefix_id", None) @@ -325,18 +321,30 @@ async def get_all_models_responses() -> list: model["id"] = f"{prefix_id}.{model['id']}" log.debug(f"get_all_models:responses() {responses}") - return responses +async def get_filtered_models(models, user): + # Filter models based on user access control + filtered_models = [] + for model in models.get("data", []): + model_info = Models.get_model_by_id(model["id"]) + if model_info: + if user.id == model_info.user_id or has_access( + user.id, type="read", access_control=model_info.access_control + ): + filtered_models.append(model) + return filtered_models + + @cached(ttl=3) -async def get_all_models() -> dict[str, list]: +async def get_all_models(request: Request) -> dict[str, list]: log.info("get_all_models()") - if not app.state.config.ENABLE_OPENAI_API: + if not request.app.state.config.ENABLE_OPENAI_API: return {"data": []} - responses = await get_all_models_responses() + responses = await get_all_models_responses(request) def extract_data(response): if response and "data" in response: @@ -345,41 +353,86 @@ async def get_all_models() -> dict[str, list]: return response return None + def merge_models_lists(model_lists): + log.debug(f"merge_models_lists {model_lists}") + merged_list = [] + + for idx, models in enumerate(model_lists): + if models is not None and "error" not in models: + merged_list.extend( + [ + { + **model, + "name": model.get("name", model["id"]), + "owned_by": "openai", + "openai": model, + "urlIdx": idx, + } + for model in models + if "api.openai.com" + not in request.app.state.config.OPENAI_API_BASE_URLS[idx] + or not any( + name in model["id"] + for name in [ + "babbage", + "dall-e", + "davinci", + "embedding", + "tts", + "whisper", + ] + ) + ] + ) + + return merged_list + models = {"data": merge_models_lists(map(extract_data, responses))} log.debug(f"models: {models}") + request.app.state.OPENAI_MODELS = {model["id"]: model for model in models["data"]} return models -@app.get("/models") -@app.get("/models/{url_idx}") -async def get_models(url_idx: Optional[int] = None, user=Depends(get_verified_user)): +@router.get("/models") +@router.get("/models/{url_idx}") +async def get_models( + request: Request, url_idx: Optional[int] = None, user=Depends(get_verified_user) +): models = { "data": [], } if url_idx is None: - models = await get_all_models() + models = await get_all_models(request) else: - url = app.state.config.OPENAI_API_BASE_URLS[url_idx] - key = app.state.config.OPENAI_API_KEYS[url_idx] - - headers = {} - headers["Authorization"] = f"Bearer {key}" - headers["Content-Type"] = "application/json" - - if ENABLE_FORWARD_USER_INFO_HEADERS: - headers["X-OpenWebUI-User-Name"] = user.name - headers["X-OpenWebUI-User-Id"] = user.id - headers["X-OpenWebUI-User-Email"] = user.email - headers["X-OpenWebUI-User-Role"] = user.role + url = request.app.state.config.OPENAI_API_BASE_URLS[url_idx] + key = request.app.state.config.OPENAI_API_KEYS[url_idx] r = None - - timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST) - async with aiohttp.ClientSession(timeout=timeout) as session: + async with aiohttp.ClientSession( + timeout=aiohttp.ClientTimeout( + total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST + ) + ) as session: try: - async with session.get(f"{url}/models", headers=headers) as r: + async with session.get( + f"{url}/models", + headers={ + "Authorization": f"Bearer {key}", + "Content-Type": "application/json", + **( + { + "X-OpenWebUI-User-Name": user.name, + "X-OpenWebUI-User-Id": user.id, + "X-OpenWebUI-User-Email": user.email, + "X-OpenWebUI-User-Role": user.role, + } + if ENABLE_FORWARD_USER_INFO_HEADERS + else {} + ), + }, + ) as r: if r.status != 200: # Extract response error details if available error_detail = f"HTTP Error: {r.status}" @@ -413,27 +466,16 @@ async def get_models(url_idx: Optional[int] = None, user=Depends(get_verified_us except aiohttp.ClientError as e: # ClientError covers all aiohttp requests issues log.exception(f"Client error: {str(e)}") - # Handle aiohttp-specific connection issues, timeout etc. raise HTTPException( status_code=500, detail="Open WebUI: Server Connection Error" ) except Exception as e: log.exception(f"Unexpected error: {e}") - # Generic error handler in case parsing JSON or other steps fail error_detail = f"Unexpected error: {str(e)}" raise HTTPException(status_code=500, detail=error_detail) if user.role == "user" and not BYPASS_MODEL_ACCESS_CONTROL: - # Filter models based on user access control - filtered_models = [] - for model in models.get("data", []): - model_info = Models.get_model_by_id(model["id"]) - if model_info: - if user.id == model_info.user_id or has_access( - user.id, type="read", access_control=model_info.access_control - ): - filtered_models.append(model) - models["data"] = filtered_models + models["data"] = get_filtered_models(models, user) return models @@ -443,21 +485,24 @@ class ConnectionVerificationForm(BaseModel): key: str -@app.post("/verify") +@router.post("/verify") async def verify_connection( form_data: ConnectionVerificationForm, user=Depends(get_admin_user) ): url = form_data.url key = form_data.key - headers = {} - headers["Authorization"] = f"Bearer {key}" - headers["Content-Type"] = "application/json" - - timeout = aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST) - async with aiohttp.ClientSession(timeout=timeout) as session: + async with aiohttp.ClientSession( + timeout=aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT_OPENAI_MODEL_LIST) + ) as session: try: - async with session.get(f"{url}/models", headers=headers) as r: + async with session.get( + f"{url}/models", + headers={ + "Authorization": f"Bearer {key}", + "Content-Type": "application/json", + }, + ) as r: if r.status != 200: # Extract response error details if available error_detail = f"HTTP Error: {r.status}" @@ -472,26 +517,24 @@ async def verify_connection( except aiohttp.ClientError as e: # ClientError covers all aiohttp requests issues log.exception(f"Client error: {str(e)}") - # Handle aiohttp-specific connection issues, timeout etc. raise HTTPException( status_code=500, detail="Open WebUI: Server Connection Error" ) except Exception as e: log.exception(f"Unexpected error: {e}") - # Generic error handler in case parsing JSON or other steps fail error_detail = f"Unexpected error: {str(e)}" raise HTTPException(status_code=500, detail=error_detail) -@app.post("/chat/completions") +@router.post("/chat/completions") async def generate_chat_completion( + request: Request, form_data: dict, user=Depends(get_verified_user), bypass_filter: Optional[bool] = False, ): idx = 0 payload = {**form_data} - if "metadata" in payload: del payload["metadata"] @@ -526,15 +569,7 @@ async def generate_chat_completion( detail="Model not found", ) - # Attemp to get urlIdx from the model - models = await get_all_models() - - # Find the model from the list - model = next( - (model for model in models["data"] if model["id"] == payload.get("model")), - None, - ) - + model = request.app.state.OPENAI_MODELS.get(model_id) if model: idx = model["urlIdx"] else: @@ -544,11 +579,11 @@ async def generate_chat_completion( ) # Get the API config for the model - api_config = app.state.config.OPENAI_API_CONFIGS.get( - app.state.config.OPENAI_API_BASE_URLS[idx], {} + api_config = request.app.state.config.OPENAI_API_CONFIGS.get( + request.app.state.config.OPENAI_API_BASE_URLS[idx], {} ) - prefix_id = api_config.get("prefix_id", None) + prefix_id = api_config.get("prefix_id", None) if prefix_id: payload["model"] = payload["model"].replace(f"{prefix_id}.", "") @@ -561,43 +596,26 @@ async def generate_chat_completion( "role": user.role, } - url = app.state.config.OPENAI_API_BASE_URLS[idx] - key = app.state.config.OPENAI_API_KEYS[idx] + url = request.app.state.config.OPENAI_API_BASE_URLS[idx] + key = request.app.state.config.OPENAI_API_KEYS[idx] # Fix: O1 does not support the "max_tokens" parameter, Modify "max_tokens" to "max_completion_tokens" is_o1 = payload["model"].lower().startswith("o1-") - # Change max_completion_tokens to max_tokens (Backward compatible) - if "api.openai.com" not in url and not is_o1: - if "max_completion_tokens" in payload: - # Remove "max_completion_tokens" from the payload - payload["max_tokens"] = payload["max_completion_tokens"] - del payload["max_completion_tokens"] - else: - if is_o1 and "max_tokens" in payload: + if is_o1: + payload = openai_o1_handler(payload) + elif "api.openai.com" not in url: + # Remove "max_tokens" from the payload for backward compatibility + if "max_tokens" in payload: payload["max_completion_tokens"] = payload["max_tokens"] del payload["max_tokens"] - if "max_tokens" in payload and "max_completion_tokens" in payload: - del payload["max_tokens"] - # Fix: O1 does not support the "system" parameter, Modify "system" to "user" - if is_o1 and payload["messages"][0]["role"] == "system": - payload["messages"][0]["role"] = "user" + # TODO: check if below is needed + # if "max_tokens" in payload and "max_completion_tokens" in payload: + # del payload["max_tokens"] # Convert the modified body back to JSON payload = json.dumps(payload) - headers = {} - headers["Authorization"] = f"Bearer {key}" - headers["Content-Type"] = "application/json" - if "openrouter.ai" in app.state.config.OPENAI_API_BASE_URLS[idx]: - headers["HTTP-Referer"] = "https://openwebui.com/" - headers["X-Title"] = "Open WebUI" - if ENABLE_FORWARD_USER_INFO_HEADERS: - headers["X-OpenWebUI-User-Name"] = user.name - headers["X-OpenWebUI-User-Id"] = user.id - headers["X-OpenWebUI-User-Email"] = user.email - headers["X-OpenWebUI-User-Role"] = user.role - r = None session = None streaming = False @@ -607,11 +625,33 @@ async def generate_chat_completion( session = aiohttp.ClientSession( trust_env=True, timeout=aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT) ) + r = await session.request( method="POST", url=f"{url}/chat/completions", data=payload, - headers=headers, + headers={ + "Authorization": f"Bearer {key}", + "Content-Type": "application/json", + **( + { + "HTTP-Referer": "https://openwebui.com/", + "X-Title": "Open WebUI", + } + if "openrouter.ai" in url + else {} + ), + **( + { + "X-OpenWebUI-User-Name": user.name, + "X-OpenWebUI-User-Id": user.id, + "X-OpenWebUI-User-Email": user.email, + "X-OpenWebUI-User-Role": user.role, + } + if ENABLE_FORWARD_USER_INFO_HEADERS + else {} + ), + }, ) # Check if response is SSE @@ -636,14 +676,18 @@ async def generate_chat_completion( return response except Exception as e: log.exception(e) - error_detail = "Open WebUI: Server Connection Error" + + detail = None if isinstance(response, dict): if "error" in response: - error_detail = f"{response['error']['message'] if 'message' in response['error'] else response['error']}" + detail = f"{response['error']['message'] if 'message' in response['error'] else response['error']}" elif isinstance(response, str): - error_detail = response + detail = response - raise HTTPException(status_code=r.status if r else 500, detail=error_detail) + raise HTTPException( + status_code=r.status if r else 500, + detail=detail if detail else "Open WebUI: Server Connection Error", + ) finally: if not streaming and session: if r: @@ -651,25 +695,17 @@ async def generate_chat_completion( await session.close() -@app.api_route("/{path:path}", methods=["GET", "POST", "PUT", "DELETE"]) +@router.api_route("/{path:path}", methods=["GET", "POST", "PUT", "DELETE"]) async def proxy(path: str, request: Request, user=Depends(get_verified_user)): - idx = 0 + """ + Deprecated: proxy all requests to OpenAI API + """ body = await request.body() - url = app.state.config.OPENAI_API_BASE_URLS[idx] - key = app.state.config.OPENAI_API_KEYS[idx] - - target_url = f"{url}/{path}" - - headers = {} - headers["Authorization"] = f"Bearer {key}" - headers["Content-Type"] = "application/json" - if ENABLE_FORWARD_USER_INFO_HEADERS: - headers["X-OpenWebUI-User-Name"] = user.name - headers["X-OpenWebUI-User-Id"] = user.id - headers["X-OpenWebUI-User-Email"] = user.email - headers["X-OpenWebUI-User-Role"] = user.role + idx = 0 + url = request.app.state.config.OPENAI_API_BASE_URLS[idx] + key = request.app.state.config.OPENAI_API_KEYS[idx] r = None session = None @@ -679,11 +715,23 @@ async def proxy(path: str, request: Request, user=Depends(get_verified_user)): session = aiohttp.ClientSession(trust_env=True) r = await session.request( method=request.method, - url=target_url, + url=f"{url}/{path}", data=body, - headers=headers, + headers={ + "Authorization": f"Bearer {key}", + "Content-Type": "application/json", + **( + { + "X-OpenWebUI-User-Name": user.name, + "X-OpenWebUI-User-Id": user.id, + "X-OpenWebUI-User-Email": user.email, + "X-OpenWebUI-User-Role": user.role, + } + if ENABLE_FORWARD_USER_INFO_HEADERS + else {} + ), + }, ) - r.raise_for_status() # Check if response is SSE @@ -700,18 +748,23 @@ async def proxy(path: str, request: Request, user=Depends(get_verified_user)): else: response_data = await r.json() return response_data + except Exception as e: log.exception(e) - error_detail = "Open WebUI: Server Connection Error" + + detail = None if r is not None: try: res = await r.json() print(res) if "error" in res: - error_detail = f"External: {res['error']['message'] if 'message' in res['error'] else res['error']}" + detail = f"External: {res['error']['message'] if 'message' in res['error'] else res['error']}" except Exception: - error_detail = f"External: {e}" - raise HTTPException(status_code=r.status if r else 500, detail=error_detail) + detail = f"External: {e}" + raise HTTPException( + status_code=r.status if r else 500, + detail=detail if detail else "Open WebUI: Server Connection Error", + ) finally: if not streaming and session: if r: diff --git a/backend/open_webui/routers/pipelines.py b/backend/open_webui/routers/pipelines.py new file mode 100644 index 000000000..258c10ee6 --- /dev/null +++ b/backend/open_webui/routers/pipelines.py @@ -0,0 +1,496 @@ +from fastapi import ( + Depends, + FastAPI, + File, + Form, + HTTPException, + Request, + UploadFile, + status, + APIRouter, +) +import os +import logging +import shutil +import requests +from pydantic import BaseModel +from starlette.responses import FileResponse +from typing import Optional + +from open_webui.env import SRC_LOG_LEVELS +from open_webui.config import CACHE_DIR +from open_webui.constants import ERROR_MESSAGES + + +from open_webui.routers.openai import get_all_models_responses + +from open_webui.utils.auth import get_admin_user + +log = logging.getLogger(__name__) +log.setLevel(SRC_LOG_LEVELS["MAIN"]) + + +################################## +# +# Pipeline Middleware +# +################################## + + +def get_sorted_filters(model_id, models): + filters = [ + model + for model in models.values() + if "pipeline" in model + and "type" in model["pipeline"] + and model["pipeline"]["type"] == "filter" + and ( + model["pipeline"]["pipelines"] == ["*"] + or any( + model_id == target_model_id + for target_model_id in model["pipeline"]["pipelines"] + ) + ) + ] + sorted_filters = sorted(filters, key=lambda x: x["pipeline"]["priority"]) + return sorted_filters + + +def process_pipeline_inlet_filter(request, payload, user, models): + user = {"id": user.id, "email": user.email, "name": user.name, "role": user.role} + model_id = payload["model"] + + sorted_filters = get_sorted_filters(model_id, models) + model = models[model_id] + + if "pipeline" in model: + sorted_filters.append(model) + + for filter in sorted_filters: + r = None + try: + urlIdx = filter["urlIdx"] + + url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx] + key = request.app.state.config.OPENAI_API_KEYS[urlIdx] + + if key == "": + continue + + headers = {"Authorization": f"Bearer {key}"} + r = requests.post( + f"{url}/{filter['id']}/filter/inlet", + headers=headers, + json={ + "user": user, + "body": payload, + }, + ) + + r.raise_for_status() + payload = r.json() + except Exception as e: + # Handle connection error here + print(f"Connection error: {e}") + + if r is not None: + res = r.json() + if "detail" in res: + raise Exception(r.status_code, res["detail"]) + + return payload + + +def process_pipeline_outlet_filter(request, payload, user, models): + user = {"id": user.id, "email": user.email, "name": user.name, "role": user.role} + model_id = payload["model"] + + sorted_filters = get_sorted_filters(model_id, models) + model = models[model_id] + + if "pipeline" in model: + sorted_filters = [model] + sorted_filters + + for filter in sorted_filters: + r = None + try: + urlIdx = filter["urlIdx"] + + url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx] + key = request.app.state.config.OPENAI_API_KEYS[urlIdx] + + if key != "": + r = requests.post( + f"{url}/{filter['id']}/filter/outlet", + headers={"Authorization": f"Bearer {key}"}, + json={ + "user": { + "id": user.id, + "name": user.name, + "email": user.email, + "role": user.role, + }, + "body": data, + }, + ) + + r.raise_for_status() + data = r.json() + except Exception as e: + # Handle connection error here + print(f"Connection error: {e}") + + if r is not None: + try: + res = r.json() + if "detail" in res: + return Exception(r.status_code, res) + except Exception: + pass + + else: + pass + + return payload + + +################################## +# +# Pipelines Endpoints +# +################################## + +router = APIRouter() + + +@router.get("/list") +async def get_pipelines_list(request: Request, user=Depends(get_admin_user)): + responses = await get_all_models_responses(request) + log.debug(f"get_pipelines_list: get_openai_models_responses returned {responses}") + + urlIdxs = [ + idx + for idx, response in enumerate(responses) + if response is not None and "pipelines" in response + ] + + return { + "data": [ + { + "url": request.app.state.config.OPENAI_API_BASE_URLS[urlIdx], + "idx": urlIdx, + } + for urlIdx in urlIdxs + ] + } + + +@router.post("/upload") +async def upload_pipeline( + request: Request, + urlIdx: int = Form(...), + file: UploadFile = File(...), + user=Depends(get_admin_user), +): + print("upload_pipeline", urlIdx, file.filename) + # Check if the uploaded file is a python file + if not (file.filename and file.filename.endswith(".py")): + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail="Only Python (.py) files are allowed.", + ) + + upload_folder = f"{CACHE_DIR}/pipelines" + os.makedirs(upload_folder, exist_ok=True) + file_path = os.path.join(upload_folder, file.filename) + + r = None + try: + # Save the uploaded file + with open(file_path, "wb") as buffer: + shutil.copyfileobj(file.file, buffer) + + url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx] + key = request.app.state.config.OPENAI_API_KEYS[urlIdx] + + with open(file_path, "rb") as f: + files = {"file": f} + r = requests.post( + f"{url}/pipelines/upload", + headers={"Authorization": f"Bearer {key}"}, + files=files, + ) + + r.raise_for_status() + data = r.json() + + return {**data} + except Exception as e: + # Handle connection error here + print(f"Connection error: {e}") + + detail = None + status_code = status.HTTP_404_NOT_FOUND + if r is not None: + status_code = r.status_code + try: + res = r.json() + if "detail" in res: + detail = res["detail"] + except Exception: + pass + + raise HTTPException( + status_code=status_code, + detail=detail if detail else "Pipeline not found", + ) + finally: + # Ensure the file is deleted after the upload is completed or on failure + if os.path.exists(file_path): + os.remove(file_path) + + +class AddPipelineForm(BaseModel): + url: str + urlIdx: int + + +@router.post("/add") +async def add_pipeline( + request: Request, form_data: AddPipelineForm, user=Depends(get_admin_user) +): + r = None + try: + urlIdx = form_data.urlIdx + + url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx] + key = request.app.state.config.OPENAI_API_KEYS[urlIdx] + + r = requests.post( + f"{url}/pipelines/add", + headers={"Authorization": f"Bearer {key}"}, + json={"url": form_data.url}, + ) + + r.raise_for_status() + data = r.json() + + return {**data} + except Exception as e: + # Handle connection error here + print(f"Connection error: {e}") + + detail = None + if r is not None: + try: + res = r.json() + if "detail" in res: + detail = res["detail"] + except Exception: + pass + + raise HTTPException( + status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), + detail=detail if detail else "Pipeline not found", + ) + + +class DeletePipelineForm(BaseModel): + id: str + urlIdx: int + + +@router.delete("/delete") +async def delete_pipeline( + request: Request, form_data: DeletePipelineForm, user=Depends(get_admin_user) +): + r = None + try: + urlIdx = form_data.urlIdx + + url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx] + key = request.app.state.config.OPENAI_API_KEYS[urlIdx] + + r = requests.delete( + f"{url}/pipelines/delete", + headers={"Authorization": f"Bearer {key}"}, + json={"id": form_data.id}, + ) + + r.raise_for_status() + data = r.json() + + return {**data} + except Exception as e: + # Handle connection error here + print(f"Connection error: {e}") + + detail = None + if r is not None: + try: + res = r.json() + if "detail" in res: + detail = res["detail"] + except Exception: + pass + + raise HTTPException( + status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), + detail=detail if detail else "Pipeline not found", + ) + + +@router.get("/") +async def get_pipelines( + request: Request, urlIdx: Optional[int] = None, user=Depends(get_admin_user) +): + r = None + try: + url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx] + key = request.app.state.config.OPENAI_API_KEYS[urlIdx] + + r = requests.get(f"{url}/pipelines", headers={"Authorization": f"Bearer {key}"}) + + r.raise_for_status() + data = r.json() + + return {**data} + except Exception as e: + # Handle connection error here + print(f"Connection error: {e}") + + detail = None + if r is not None: + try: + res = r.json() + if "detail" in res: + detail = res["detail"] + except Exception: + pass + + raise HTTPException( + status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), + detail=detail if detail else "Pipeline not found", + ) + + +@router.get("/{pipeline_id}/valves") +async def get_pipeline_valves( + request: Request, + urlIdx: Optional[int], + pipeline_id: str, + user=Depends(get_admin_user), +): + r = None + try: + url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx] + key = request.app.state.config.OPENAI_API_KEYS[urlIdx] + + r = requests.get( + f"{url}/{pipeline_id}/valves", headers={"Authorization": f"Bearer {key}"} + ) + + r.raise_for_status() + data = r.json() + + return {**data} + except Exception as e: + # Handle connection error here + print(f"Connection error: {e}") + + detail = None + if r is not None: + try: + res = r.json() + if "detail" in res: + detail = res["detail"] + except Exception: + pass + + raise HTTPException( + status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), + detail=detail if detail else "Pipeline not found", + ) + + +@router.get("/{pipeline_id}/valves/spec") +async def get_pipeline_valves_spec( + request: Request, + urlIdx: Optional[int], + pipeline_id: str, + user=Depends(get_admin_user), +): + r = None + try: + url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx] + key = request.app.state.config.OPENAI_API_KEYS[urlIdx] + + r = requests.get( + f"{url}/{pipeline_id}/valves/spec", + headers={"Authorization": f"Bearer {key}"}, + ) + + r.raise_for_status() + data = r.json() + + return {**data} + except Exception as e: + # Handle connection error here + print(f"Connection error: {e}") + + detail = None + if r is not None: + try: + res = r.json() + if "detail" in res: + detail = res["detail"] + except Exception: + pass + + raise HTTPException( + status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), + detail=detail if detail else "Pipeline not found", + ) + + +@router.post("/{pipeline_id}/valves/update") +async def update_pipeline_valves( + request: Request, + urlIdx: Optional[int], + pipeline_id: str, + form_data: dict, + user=Depends(get_admin_user), +): + r = None + try: + url = request.app.state.config.OPENAI_API_BASE_URLS[urlIdx] + key = request.app.state.config.OPENAI_API_KEYS[urlIdx] + + r = requests.post( + f"{url}/{pipeline_id}/valves/update", + headers={"Authorization": f"Bearer {key}"}, + json={**form_data}, + ) + + r.raise_for_status() + data = r.json() + + return {**data} + except Exception as e: + # Handle connection error here + print(f"Connection error: {e}") + + detail = None + + if r is not None: + try: + res = r.json() + if "detail" in res: + detail = res["detail"] + except Exception: + pass + + raise HTTPException( + status_code=(r.status_code if r is not None else status.HTTP_404_NOT_FOUND), + detail=detail if detail else "Pipeline not found", + ) diff --git a/backend/open_webui/apps/webui/routers/prompts.py b/backend/open_webui/routers/prompts.py similarity index 98% rename from backend/open_webui/apps/webui/routers/prompts.py rename to backend/open_webui/routers/prompts.py index 89a60fd95..4f1c48482 100644 --- a/backend/open_webui/apps/webui/routers/prompts.py +++ b/backend/open_webui/routers/prompts.py @@ -1,6 +1,6 @@ from typing import Optional -from open_webui.apps.webui.models.prompts import ( +from open_webui.models.prompts import ( PromptForm, PromptUserResponse, PromptModel, diff --git a/backend/open_webui/apps/retrieval/main.py b/backend/open_webui/routers/retrieval.py similarity index 52% rename from backend/open_webui/apps/retrieval/main.py rename to backend/open_webui/routers/retrieval.py index 4da322e70..1898bfe49 100644 --- a/backend/open_webui/apps/retrieval/main.py +++ b/backend/open_webui/routers/retrieval.py @@ -1,47 +1,63 @@ -# TODO: Merge this with the webui_app and make it a single app - import json import logging +import mimetypes import os import shutil import uuid from datetime import datetime -from typing import List, Optional +from pathlib import Path +from typing import Iterator, List, Optional, Sequence, Union -from fastapi import Depends, FastAPI, HTTPException, status +from fastapi import ( + Depends, + FastAPI, + File, + Form, + HTTPException, + UploadFile, + Request, + status, + APIRouter, +) from fastapi.middleware.cors import CORSMiddleware from pydantic import BaseModel import tiktoken +from langchain.text_splitter import RecursiveCharacterTextSplitter, TokenTextSplitter +from langchain_core.documents import Document + +from open_webui.models.files import FileModel, Files +from open_webui.models.knowledge import Knowledges from open_webui.storage.provider import Storage -from open_webui.apps.webui.models.knowledge import Knowledges -from open_webui.apps.retrieval.vector.connector import VECTOR_DB_CLIENT + + +from open_webui.retrieval.vector.connector import VECTOR_DB_CLIENT # Document loaders -from open_webui.apps.retrieval.loaders.main import Loader -from open_webui.apps.retrieval.loaders.youtube import YoutubeLoader +from open_webui.retrieval.loaders.main import Loader +from open_webui.retrieval.loaders.youtube import YoutubeLoader # Web search engines -from open_webui.apps.retrieval.web.main import SearchResult -from open_webui.apps.retrieval.web.utils import get_web_loader -from open_webui.apps.retrieval.web.brave import search_brave -from open_webui.apps.retrieval.web.kagi import search_kagi -from open_webui.apps.retrieval.web.mojeek import search_mojeek -from open_webui.apps.retrieval.web.duckduckgo import search_duckduckgo -from open_webui.apps.retrieval.web.google_pse import search_google_pse -from open_webui.apps.retrieval.web.jina_search import search_jina -from open_webui.apps.retrieval.web.searchapi import search_searchapi -from open_webui.apps.retrieval.web.searxng import search_searxng -from open_webui.apps.retrieval.web.serper import search_serper -from open_webui.apps.retrieval.web.serply import search_serply -from open_webui.apps.retrieval.web.serpstack import search_serpstack -from open_webui.apps.retrieval.web.tavily import search_tavily -from open_webui.apps.retrieval.web.bing import search_bing +from open_webui.retrieval.web.main import SearchResult +from open_webui.retrieval.web.utils import get_web_loader +from open_webui.retrieval.web.brave import search_brave +from open_webui.retrieval.web.kagi import search_kagi +from open_webui.retrieval.web.mojeek import search_mojeek +from open_webui.retrieval.web.duckduckgo import search_duckduckgo +from open_webui.retrieval.web.google_pse import search_google_pse +from open_webui.retrieval.web.jina_search import search_jina +from open_webui.retrieval.web.searchapi import search_searchapi +from open_webui.retrieval.web.searxng import search_searxng +from open_webui.retrieval.web.serper import search_serper +from open_webui.retrieval.web.serply import search_serply +from open_webui.retrieval.web.serpstack import search_serpstack +from open_webui.retrieval.web.tavily import search_tavily +from open_webui.retrieval.web.bing import search_bing -from open_webui.apps.retrieval.utils import ( +from open_webui.retrieval.utils import ( get_embedding_function, get_model_path, query_collection, @@ -49,242 +65,100 @@ from open_webui.apps.retrieval.utils import ( query_doc, query_doc_with_hybrid_search, ) - -from open_webui.apps.webui.models.files import FileModel, Files -from open_webui.config import ( - BRAVE_SEARCH_API_KEY, - KAGI_SEARCH_API_KEY, - MOJEEK_SEARCH_API_KEY, - TIKTOKEN_ENCODING_NAME, - RAG_TEXT_SPLITTER, - CHUNK_OVERLAP, - CHUNK_SIZE, - CONTENT_EXTRACTION_ENGINE, - CORS_ALLOW_ORIGIN, - ENABLE_RAG_HYBRID_SEARCH, - ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION, - ENABLE_RAG_WEB_SEARCH, - ENV, - GOOGLE_PSE_API_KEY, - GOOGLE_PSE_ENGINE_ID, - PDF_EXTRACT_IMAGES, - RAG_EMBEDDING_ENGINE, - RAG_EMBEDDING_MODEL, - RAG_EMBEDDING_MODEL_AUTO_UPDATE, - RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE, - RAG_EMBEDDING_BATCH_SIZE, - RAG_FILE_MAX_COUNT, - RAG_FILE_MAX_SIZE, - RAG_OPENAI_API_BASE_URL, - RAG_OPENAI_API_KEY, - RAG_OLLAMA_BASE_URL, - RAG_OLLAMA_API_KEY, - RAG_RELEVANCE_THRESHOLD, - RAG_RERANKING_MODEL, - RAG_RERANKING_MODEL_AUTO_UPDATE, - RAG_RERANKING_MODEL_TRUST_REMOTE_CODE, - RAG_TEMPLATE, - RAG_TOP_K, - RAG_WEB_SEARCH_CONCURRENT_REQUESTS, - RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - RAG_WEB_SEARCH_ENGINE, - RAG_WEB_SEARCH_RESULT_COUNT, - JINA_API_KEY, - SEARCHAPI_API_KEY, - SEARCHAPI_ENGINE, - SEARXNG_QUERY_URL, - SERPER_API_KEY, - SERPLY_API_KEY, - SERPSTACK_API_KEY, - SERPSTACK_HTTPS, - TAVILY_API_KEY, - BING_SEARCH_V7_ENDPOINT, - BING_SEARCH_V7_SUBSCRIPTION_KEY, - TIKA_SERVER_URL, - UPLOAD_DIR, - YOUTUBE_LOADER_LANGUAGE, - YOUTUBE_LOADER_PROXY_URL, - DEFAULT_LOCALE, - AppConfig, -) -from open_webui.constants import ERROR_MESSAGES -from open_webui.env import ( - SRC_LOG_LEVELS, - DEVICE_TYPE, - DOCKER, -) from open_webui.utils.misc import ( calculate_sha256_string, ) from open_webui.utils.auth import get_admin_user, get_verified_user -from langchain.text_splitter import RecursiveCharacterTextSplitter, TokenTextSplitter -from langchain_core.documents import Document +from open_webui.config import ( + ENV, + RAG_EMBEDDING_MODEL_AUTO_UPDATE, + RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE, + RAG_RERANKING_MODEL_AUTO_UPDATE, + RAG_RERANKING_MODEL_TRUST_REMOTE_CODE, + UPLOAD_DIR, + DEFAULT_LOCALE, +) +from open_webui.env import ( + SRC_LOG_LEVELS, + DEVICE_TYPE, + DOCKER, +) +from open_webui.constants import ERROR_MESSAGES log = logging.getLogger(__name__) log.setLevel(SRC_LOG_LEVELS["RAG"]) -app = FastAPI( - docs_url="/docs" if ENV == "dev" else None, - openapi_url="/openapi.json" if ENV == "dev" else None, - redoc_url=None, -) - -app.state.config = AppConfig() - -app.state.config.TOP_K = RAG_TOP_K -app.state.config.RELEVANCE_THRESHOLD = RAG_RELEVANCE_THRESHOLD -app.state.config.FILE_MAX_SIZE = RAG_FILE_MAX_SIZE -app.state.config.FILE_MAX_COUNT = RAG_FILE_MAX_COUNT - -app.state.config.ENABLE_RAG_HYBRID_SEARCH = ENABLE_RAG_HYBRID_SEARCH -app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION = ( - ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION -) - -app.state.config.CONTENT_EXTRACTION_ENGINE = CONTENT_EXTRACTION_ENGINE -app.state.config.TIKA_SERVER_URL = TIKA_SERVER_URL - -app.state.config.TEXT_SPLITTER = RAG_TEXT_SPLITTER -app.state.config.TIKTOKEN_ENCODING_NAME = TIKTOKEN_ENCODING_NAME - -app.state.config.CHUNK_SIZE = CHUNK_SIZE -app.state.config.CHUNK_OVERLAP = CHUNK_OVERLAP - -app.state.config.RAG_EMBEDDING_ENGINE = RAG_EMBEDDING_ENGINE -app.state.config.RAG_EMBEDDING_MODEL = RAG_EMBEDDING_MODEL -app.state.config.RAG_EMBEDDING_BATCH_SIZE = RAG_EMBEDDING_BATCH_SIZE -app.state.config.RAG_RERANKING_MODEL = RAG_RERANKING_MODEL -app.state.config.RAG_TEMPLATE = RAG_TEMPLATE - -app.state.config.OPENAI_API_BASE_URL = RAG_OPENAI_API_BASE_URL -app.state.config.OPENAI_API_KEY = RAG_OPENAI_API_KEY - -app.state.config.OLLAMA_BASE_URL = RAG_OLLAMA_BASE_URL -app.state.config.OLLAMA_API_KEY = RAG_OLLAMA_API_KEY - -app.state.config.PDF_EXTRACT_IMAGES = PDF_EXTRACT_IMAGES - -app.state.config.YOUTUBE_LOADER_LANGUAGE = YOUTUBE_LOADER_LANGUAGE -app.state.config.YOUTUBE_LOADER_PROXY_URL = YOUTUBE_LOADER_PROXY_URL -app.state.YOUTUBE_LOADER_TRANSLATION = None +########################################## +# +# Utility functions +# +########################################## -app.state.config.ENABLE_RAG_WEB_SEARCH = ENABLE_RAG_WEB_SEARCH -app.state.config.RAG_WEB_SEARCH_ENGINE = RAG_WEB_SEARCH_ENGINE -app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST = RAG_WEB_SEARCH_DOMAIN_FILTER_LIST - -app.state.config.SEARXNG_QUERY_URL = SEARXNG_QUERY_URL -app.state.config.GOOGLE_PSE_API_KEY = GOOGLE_PSE_API_KEY -app.state.config.GOOGLE_PSE_ENGINE_ID = GOOGLE_PSE_ENGINE_ID -app.state.config.BRAVE_SEARCH_API_KEY = BRAVE_SEARCH_API_KEY -app.state.config.KAGI_SEARCH_API_KEY = KAGI_SEARCH_API_KEY -app.state.config.MOJEEK_SEARCH_API_KEY = MOJEEK_SEARCH_API_KEY -app.state.config.SERPSTACK_API_KEY = SERPSTACK_API_KEY -app.state.config.SERPSTACK_HTTPS = SERPSTACK_HTTPS -app.state.config.SERPER_API_KEY = SERPER_API_KEY -app.state.config.SERPLY_API_KEY = SERPLY_API_KEY -app.state.config.TAVILY_API_KEY = TAVILY_API_KEY -app.state.config.SEARCHAPI_API_KEY = SEARCHAPI_API_KEY -app.state.config.SEARCHAPI_ENGINE = SEARCHAPI_ENGINE -app.state.config.JINA_API_KEY = JINA_API_KEY -app.state.config.BING_SEARCH_V7_ENDPOINT = BING_SEARCH_V7_ENDPOINT -app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY = BING_SEARCH_V7_SUBSCRIPTION_KEY - -app.state.config.RAG_WEB_SEARCH_RESULT_COUNT = RAG_WEB_SEARCH_RESULT_COUNT -app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS = RAG_WEB_SEARCH_CONCURRENT_REQUESTS - - -def update_embedding_model( +def get_ef( + engine: str, embedding_model: str, auto_update: bool = False, ): - if embedding_model and app.state.config.RAG_EMBEDDING_ENGINE == "": + ef = None + if embedding_model and engine == "": from sentence_transformers import SentenceTransformer try: - app.state.sentence_transformer_ef = SentenceTransformer( + ef = SentenceTransformer( get_model_path(embedding_model, auto_update), device=DEVICE_TYPE, trust_remote_code=RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE, ) except Exception as e: log.debug(f"Error loading SentenceTransformer: {e}") - app.state.sentence_transformer_ef = None - else: - app.state.sentence_transformer_ef = None + + return ef -def update_reranking_model( +def get_rf( reranking_model: str, auto_update: bool = False, ): + rf = None if reranking_model: if any(model in reranking_model for model in ["jinaai/jina-colbert-v2"]): try: - from open_webui.apps.retrieval.models.colbert import ColBERT + from open_webui.retrieval.models.colbert import ColBERT - app.state.sentence_transformer_rf = ColBERT( + rf = ColBERT( get_model_path(reranking_model, auto_update), env="docker" if DOCKER else None, ) + except Exception as e: log.error(f"ColBERT: {e}") - app.state.sentence_transformer_rf = None - app.state.config.ENABLE_RAG_HYBRID_SEARCH = False + raise Exception(ERROR_MESSAGES.DEFAULT(e)) else: import sentence_transformers try: - app.state.sentence_transformer_rf = sentence_transformers.CrossEncoder( + rf = sentence_transformers.CrossEncoder( get_model_path(reranking_model, auto_update), device=DEVICE_TYPE, trust_remote_code=RAG_RERANKING_MODEL_TRUST_REMOTE_CODE, ) except: log.error("CrossEncoder error") - app.state.sentence_transformer_rf = None - app.state.config.ENABLE_RAG_HYBRID_SEARCH = False - else: - app.state.sentence_transformer_rf = None + raise Exception(ERROR_MESSAGES.DEFAULT("CrossEncoder error")) + return rf -update_embedding_model( - app.state.config.RAG_EMBEDDING_MODEL, - RAG_EMBEDDING_MODEL_AUTO_UPDATE, -) - -update_reranking_model( - app.state.config.RAG_RERANKING_MODEL, - RAG_RERANKING_MODEL_AUTO_UPDATE, -) +########################################## +# +# API routes +# +########################################## -app.state.EMBEDDING_FUNCTION = get_embedding_function( - app.state.config.RAG_EMBEDDING_ENGINE, - app.state.config.RAG_EMBEDDING_MODEL, - app.state.sentence_transformer_ef, - ( - app.state.config.OPENAI_API_BASE_URL - if app.state.config.RAG_EMBEDDING_ENGINE == "openai" - else app.state.config.OLLAMA_BASE_URL - ), - ( - app.state.config.OPENAI_API_KEY - if app.state.config.RAG_EMBEDDING_ENGINE == "openai" - else app.state.config.OLLAMA_API_KEY - ), - app.state.config.RAG_EMBEDDING_BATCH_SIZE, -) - -app.add_middleware( - CORSMiddleware, - allow_origins=CORS_ALLOW_ORIGIN, - allow_credentials=True, - allow_methods=["*"], - allow_headers=["*"], -) +router = APIRouter() class CollectionNameForm(BaseModel): @@ -299,43 +173,43 @@ class SearchForm(CollectionNameForm): query: str -@app.get("/") -async def get_status(): +@router.get("/") +async def get_status(request: Request): return { "status": True, - "chunk_size": app.state.config.CHUNK_SIZE, - "chunk_overlap": app.state.config.CHUNK_OVERLAP, - "template": app.state.config.RAG_TEMPLATE, - "embedding_engine": app.state.config.RAG_EMBEDDING_ENGINE, - "embedding_model": app.state.config.RAG_EMBEDDING_MODEL, - "reranking_model": app.state.config.RAG_RERANKING_MODEL, - "embedding_batch_size": app.state.config.RAG_EMBEDDING_BATCH_SIZE, + "chunk_size": request.app.state.config.CHUNK_SIZE, + "chunk_overlap": request.app.state.config.CHUNK_OVERLAP, + "template": request.app.state.config.RAG_TEMPLATE, + "embedding_engine": request.app.state.config.RAG_EMBEDDING_ENGINE, + "embedding_model": request.app.state.config.RAG_EMBEDDING_MODEL, + "reranking_model": request.app.state.config.RAG_RERANKING_MODEL, + "embedding_batch_size": request.app.state.config.RAG_EMBEDDING_BATCH_SIZE, } -@app.get("/embedding") -async def get_embedding_config(user=Depends(get_admin_user)): +@router.get("/embedding") +async def get_embedding_config(request: Request, user=Depends(get_admin_user)): return { "status": True, - "embedding_engine": app.state.config.RAG_EMBEDDING_ENGINE, - "embedding_model": app.state.config.RAG_EMBEDDING_MODEL, - "embedding_batch_size": app.state.config.RAG_EMBEDDING_BATCH_SIZE, + "embedding_engine": request.app.state.config.RAG_EMBEDDING_ENGINE, + "embedding_model": request.app.state.config.RAG_EMBEDDING_MODEL, + "embedding_batch_size": request.app.state.config.RAG_EMBEDDING_BATCH_SIZE, "openai_config": { - "url": app.state.config.OPENAI_API_BASE_URL, - "key": app.state.config.OPENAI_API_KEY, + "url": request.app.state.config.RAG_OPENAI_API_BASE_URL, + "key": request.app.state.config.RAG_OPENAI_API_KEY, }, "ollama_config": { - "url": app.state.config.OLLAMA_BASE_URL, - "key": app.state.config.OLLAMA_API_KEY, + "url": request.app.state.config.RAG_OLLAMA_BASE_URL, + "key": request.app.state.config.RAG_OLLAMA_API_KEY, }, } -@app.get("/reranking") -async def get_reraanking_config(user=Depends(get_admin_user)): +@router.get("/reranking") +async def get_reraanking_config(request: Request, user=Depends(get_admin_user)): return { "status": True, - "reranking_model": app.state.config.RAG_RERANKING_MODEL, + "reranking_model": request.app.state.config.RAG_RERANKING_MODEL, } @@ -357,59 +231,72 @@ class EmbeddingModelUpdateForm(BaseModel): embedding_batch_size: Optional[int] = 1 -@app.post("/embedding/update") +@router.post("/embedding/update") async def update_embedding_config( - form_data: EmbeddingModelUpdateForm, user=Depends(get_admin_user) + request: Request, form_data: EmbeddingModelUpdateForm, user=Depends(get_admin_user) ): log.info( - f"Updating embedding model: {app.state.config.RAG_EMBEDDING_MODEL} to {form_data.embedding_model}" + f"Updating embedding model: {request.app.state.config.RAG_EMBEDDING_MODEL} to {form_data.embedding_model}" ) try: - app.state.config.RAG_EMBEDDING_ENGINE = form_data.embedding_engine - app.state.config.RAG_EMBEDDING_MODEL = form_data.embedding_model + request.app.state.config.RAG_EMBEDDING_ENGINE = form_data.embedding_engine + request.app.state.config.RAG_EMBEDDING_MODEL = form_data.embedding_model - if app.state.config.RAG_EMBEDDING_ENGINE in ["ollama", "openai"]: + if request.app.state.config.RAG_EMBEDDING_ENGINE in ["ollama", "openai"]: if form_data.openai_config is not None: - app.state.config.OPENAI_API_BASE_URL = form_data.openai_config.url - app.state.config.OPENAI_API_KEY = form_data.openai_config.key + request.app.state.config.RAG_OPENAI_API_BASE_URL = ( + form_data.openai_config.url + ) + request.app.state.config.RAG_OPENAI_API_KEY = ( + form_data.openai_config.key + ) if form_data.ollama_config is not None: - app.state.config.OLLAMA_BASE_URL = form_data.ollama_config.url - app.state.config.OLLAMA_API_KEY = form_data.ollama_config.key + request.app.state.config.RAG_OLLAMA_BASE_URL = ( + form_data.ollama_config.url + ) + request.app.state.config.RAG_OLLAMA_API_KEY = ( + form_data.ollama_config.key + ) - app.state.config.RAG_EMBEDDING_BATCH_SIZE = form_data.embedding_batch_size + request.app.state.config.RAG_EMBEDDING_BATCH_SIZE = ( + form_data.embedding_batch_size + ) - update_embedding_model(app.state.config.RAG_EMBEDDING_MODEL) + request.app.state.ef = get_ef( + request.app.state.config.RAG_EMBEDDING_ENGINE, + request.app.state.config.RAG_EMBEDDING_MODEL, + ) - app.state.EMBEDDING_FUNCTION = get_embedding_function( - app.state.config.RAG_EMBEDDING_ENGINE, - app.state.config.RAG_EMBEDDING_MODEL, - app.state.sentence_transformer_ef, + request.app.state.EMBEDDING_FUNCTION = get_embedding_function( + request.app.state.config.RAG_EMBEDDING_ENGINE, + request.app.state.config.RAG_EMBEDDING_MODEL, + request.app.state.ef, ( - app.state.config.OPENAI_API_BASE_URL - if app.state.config.RAG_EMBEDDING_ENGINE == "openai" - else app.state.config.OLLAMA_BASE_URL + request.app.state.config.RAG_OPENAI_API_BASE_URL + if request.app.state.config.RAG_EMBEDDING_ENGINE == "openai" + else request.app.state.config.RAG_OLLAMA_BASE_URL ), ( - app.state.config.OPENAI_API_KEY - if app.state.config.RAG_EMBEDDING_ENGINE == "openai" - else app.state.config.OLLAMA_API_KEY + request.app.state.config.RAG_OPENAI_API_KEY + if request.app.state.config.RAG_EMBEDDING_ENGINE == "openai" + else request.app.state.config.RAG_OLLAMA_API_KEY ), - app.state.config.RAG_EMBEDDING_BATCH_SIZE, + request.app.state.config.RAG_EMBEDDING_BATCH_SIZE, ) return { "status": True, - "embedding_engine": app.state.config.RAG_EMBEDDING_ENGINE, - "embedding_model": app.state.config.RAG_EMBEDDING_MODEL, - "embedding_batch_size": app.state.config.RAG_EMBEDDING_BATCH_SIZE, + "embedding_engine": request.app.state.config.RAG_EMBEDDING_ENGINE, + "embedding_model": request.app.state.config.RAG_EMBEDDING_MODEL, + "embedding_batch_size": request.app.state.config.RAG_EMBEDDING_BATCH_SIZE, "openai_config": { - "url": app.state.config.OPENAI_API_BASE_URL, - "key": app.state.config.OPENAI_API_KEY, + "url": request.app.state.config.RAG_OPENAI_API_BASE_URL, + "key": request.app.state.config.RAG_OPENAI_API_KEY, }, "ollama_config": { - "url": app.state.config.OLLAMA_BASE_URL, - "key": app.state.config.OLLAMA_API_KEY, + "url": request.app.state.config.RAG_OLLAMA_BASE_URL, + "key": request.app.state.config.RAG_OLLAMA_API_KEY, }, } except Exception as e: @@ -424,21 +311,28 @@ class RerankingModelUpdateForm(BaseModel): reranking_model: str -@app.post("/reranking/update") +@router.post("/reranking/update") async def update_reranking_config( - form_data: RerankingModelUpdateForm, user=Depends(get_admin_user) + request: Request, form_data: RerankingModelUpdateForm, user=Depends(get_admin_user) ): log.info( - f"Updating reranking model: {app.state.config.RAG_RERANKING_MODEL} to {form_data.reranking_model}" + f"Updating reranking model: {request.app.state.config.RAG_RERANKING_MODEL} to {form_data.reranking_model}" ) try: - app.state.config.RAG_RERANKING_MODEL = form_data.reranking_model + request.app.state.config.RAG_RERANKING_MODEL = form_data.reranking_model - update_reranking_model(app.state.config.RAG_RERANKING_MODEL, True) + try: + request.app.state.rf = get_rf( + request.app.state.config.RAG_RERANKING_MODEL, + True, + ) + except Exception as e: + log.error(f"Error loading reranking model: {e}") + request.app.state.config.ENABLE_RAG_HYBRID_SEARCH = False return { "status": True, - "reranking_model": app.state.config.RAG_RERANKING_MODEL, + "reranking_model": request.app.state.config.RAG_RERANKING_MODEL, } except Exception as e: log.exception(f"Problem updating reranking model: {e}") @@ -448,52 +342,52 @@ async def update_reranking_config( ) -@app.get("/config") -async def get_rag_config(user=Depends(get_admin_user)): +@router.get("/config") +async def get_rag_config(request: Request, user=Depends(get_admin_user)): return { "status": True, - "pdf_extract_images": app.state.config.PDF_EXTRACT_IMAGES, + "pdf_extract_images": request.app.state.config.PDF_EXTRACT_IMAGES, "content_extraction": { - "engine": app.state.config.CONTENT_EXTRACTION_ENGINE, - "tika_server_url": app.state.config.TIKA_SERVER_URL, + "engine": request.app.state.config.CONTENT_EXTRACTION_ENGINE, + "tika_server_url": request.app.state.config.TIKA_SERVER_URL, }, "chunk": { - "text_splitter": app.state.config.TEXT_SPLITTER, - "chunk_size": app.state.config.CHUNK_SIZE, - "chunk_overlap": app.state.config.CHUNK_OVERLAP, + "text_splitter": request.app.state.config.TEXT_SPLITTER, + "chunk_size": request.app.state.config.CHUNK_SIZE, + "chunk_overlap": request.app.state.config.CHUNK_OVERLAP, }, "file": { - "max_size": app.state.config.FILE_MAX_SIZE, - "max_count": app.state.config.FILE_MAX_COUNT, + "max_size": request.app.state.config.FILE_MAX_SIZE, + "max_count": request.app.state.config.FILE_MAX_COUNT, }, "youtube": { - "language": app.state.config.YOUTUBE_LOADER_LANGUAGE, - "translation": app.state.YOUTUBE_LOADER_TRANSLATION, - "proxy_url": app.state.config.YOUTUBE_LOADER_PROXY_URL, + "language": request.app.state.config.YOUTUBE_LOADER_LANGUAGE, + "translation": request.app.state.YOUTUBE_LOADER_TRANSLATION, + "proxy_url": request.app.state.config.YOUTUBE_LOADER_PROXY_URL, }, "web": { - "web_loader_ssl_verification": app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION, + "web_loader_ssl_verification": request.app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION, "search": { - "enabled": app.state.config.ENABLE_RAG_WEB_SEARCH, - "engine": app.state.config.RAG_WEB_SEARCH_ENGINE, - "searxng_query_url": app.state.config.SEARXNG_QUERY_URL, - "google_pse_api_key": app.state.config.GOOGLE_PSE_API_KEY, - "google_pse_engine_id": app.state.config.GOOGLE_PSE_ENGINE_ID, - "brave_search_api_key": app.state.config.BRAVE_SEARCH_API_KEY, - "kagi_search_api_key": app.state.config.KAGI_SEARCH_API_KEY, - "mojeek_search_api_key": app.state.config.MOJEEK_SEARCH_API_KEY, - "serpstack_api_key": app.state.config.SERPSTACK_API_KEY, - "serpstack_https": app.state.config.SERPSTACK_HTTPS, - "serper_api_key": app.state.config.SERPER_API_KEY, - "serply_api_key": app.state.config.SERPLY_API_KEY, - "tavily_api_key": app.state.config.TAVILY_API_KEY, - "searchapi_api_key": app.state.config.SEARCHAPI_API_KEY, - "seaarchapi_engine": app.state.config.SEARCHAPI_ENGINE, - "jina_api_key": app.state.config.JINA_API_KEY, - "bing_search_v7_endpoint": app.state.config.BING_SEARCH_V7_ENDPOINT, - "bing_search_v7_subscription_key": app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY, - "result_count": app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - "concurrent_requests": app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS, + "enabled": request.app.state.config.ENABLE_RAG_WEB_SEARCH, + "engine": request.app.state.config.RAG_WEB_SEARCH_ENGINE, + "searxng_query_url": request.app.state.config.SEARXNG_QUERY_URL, + "google_pse_api_key": request.app.state.config.GOOGLE_PSE_API_KEY, + "google_pse_engine_id": request.app.state.config.GOOGLE_PSE_ENGINE_ID, + "brave_search_api_key": request.app.state.config.BRAVE_SEARCH_API_KEY, + "kagi_search_api_key": request.app.state.config.KAGI_SEARCH_API_KEY, + "mojeek_search_api_key": request.app.state.config.MOJEEK_SEARCH_API_KEY, + "serpstack_api_key": request.app.state.config.SERPSTACK_API_KEY, + "serpstack_https": request.app.state.config.SERPSTACK_HTTPS, + "serper_api_key": request.app.state.config.SERPER_API_KEY, + "serply_api_key": request.app.state.config.SERPLY_API_KEY, + "tavily_api_key": request.app.state.config.TAVILY_API_KEY, + "searchapi_api_key": request.app.state.config.SEARCHAPI_API_KEY, + "seaarchapi_engine": request.app.state.config.SEARCHAPI_ENGINE, + "jina_api_key": request.app.state.config.JINA_API_KEY, + "bing_search_v7_endpoint": request.app.state.config.BING_SEARCH_V7_ENDPOINT, + "bing_search_v7_subscription_key": request.app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY, + "result_count": request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + "concurrent_requests": request.app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS, }, }, } @@ -558,139 +452,159 @@ class ConfigUpdateForm(BaseModel): web: Optional[WebConfig] = None -@app.post("/config/update") -async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_user)): - app.state.config.PDF_EXTRACT_IMAGES = ( +@router.post("/config/update") +async def update_rag_config( + request: Request, form_data: ConfigUpdateForm, user=Depends(get_admin_user) +): + request.app.state.config.PDF_EXTRACT_IMAGES = ( form_data.pdf_extract_images if form_data.pdf_extract_images is not None - else app.state.config.PDF_EXTRACT_IMAGES + else request.app.state.config.PDF_EXTRACT_IMAGES ) if form_data.file is not None: - app.state.config.FILE_MAX_SIZE = form_data.file.max_size - app.state.config.FILE_MAX_COUNT = form_data.file.max_count + request.app.state.config.FILE_MAX_SIZE = form_data.file.max_size + request.app.state.config.FILE_MAX_COUNT = form_data.file.max_count if form_data.content_extraction is not None: log.info(f"Updating text settings: {form_data.content_extraction}") - app.state.config.CONTENT_EXTRACTION_ENGINE = form_data.content_extraction.engine - app.state.config.TIKA_SERVER_URL = form_data.content_extraction.tika_server_url + request.app.state.config.CONTENT_EXTRACTION_ENGINE = ( + form_data.content_extraction.engine + ) + request.app.state.config.TIKA_SERVER_URL = ( + form_data.content_extraction.tika_server_url + ) if form_data.chunk is not None: - app.state.config.TEXT_SPLITTER = form_data.chunk.text_splitter - app.state.config.CHUNK_SIZE = form_data.chunk.chunk_size - app.state.config.CHUNK_OVERLAP = form_data.chunk.chunk_overlap + request.app.state.config.TEXT_SPLITTER = form_data.chunk.text_splitter + request.app.state.config.CHUNK_SIZE = form_data.chunk.chunk_size + request.app.state.config.CHUNK_OVERLAP = form_data.chunk.chunk_overlap if form_data.youtube is not None: - app.state.config.YOUTUBE_LOADER_LANGUAGE = form_data.youtube.language - app.state.config.YOUTUBE_LOADER_PROXY_URL = form_data.youtube.proxy_url - app.state.YOUTUBE_LOADER_TRANSLATION = form_data.youtube.translation + request.app.state.config.YOUTUBE_LOADER_LANGUAGE = form_data.youtube.language + request.app.state.config.YOUTUBE_LOADER_PROXY_URL = form_data.youtube.proxy_url + request.app.state.YOUTUBE_LOADER_TRANSLATION = form_data.youtube.translation if form_data.web is not None: - app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION = ( + request.app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION = ( # Note: When UI "Bypass SSL verification for Websites"=True then ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION=False form_data.web.web_loader_ssl_verification ) - app.state.config.ENABLE_RAG_WEB_SEARCH = form_data.web.search.enabled - app.state.config.RAG_WEB_SEARCH_ENGINE = form_data.web.search.engine - app.state.config.SEARXNG_QUERY_URL = form_data.web.search.searxng_query_url - app.state.config.GOOGLE_PSE_API_KEY = form_data.web.search.google_pse_api_key - app.state.config.GOOGLE_PSE_ENGINE_ID = ( + request.app.state.config.ENABLE_RAG_WEB_SEARCH = form_data.web.search.enabled + request.app.state.config.RAG_WEB_SEARCH_ENGINE = form_data.web.search.engine + request.app.state.config.SEARXNG_QUERY_URL = ( + form_data.web.search.searxng_query_url + ) + request.app.state.config.GOOGLE_PSE_API_KEY = ( + form_data.web.search.google_pse_api_key + ) + request.app.state.config.GOOGLE_PSE_ENGINE_ID = ( form_data.web.search.google_pse_engine_id ) - app.state.config.BRAVE_SEARCH_API_KEY = ( + request.app.state.config.BRAVE_SEARCH_API_KEY = ( form_data.web.search.brave_search_api_key ) - app.state.config.KAGI_SEARCH_API_KEY = form_data.web.search.kagi_search_api_key - app.state.config.MOJEEK_SEARCH_API_KEY = ( + request.app.state.config.KAGI_SEARCH_API_KEY = ( + form_data.web.search.kagi_search_api_key + ) + request.app.state.config.MOJEEK_SEARCH_API_KEY = ( form_data.web.search.mojeek_search_api_key ) - app.state.config.SERPSTACK_API_KEY = form_data.web.search.serpstack_api_key - app.state.config.SERPSTACK_HTTPS = form_data.web.search.serpstack_https - app.state.config.SERPER_API_KEY = form_data.web.search.serper_api_key - app.state.config.SERPLY_API_KEY = form_data.web.search.serply_api_key - app.state.config.TAVILY_API_KEY = form_data.web.search.tavily_api_key - app.state.config.SEARCHAPI_API_KEY = form_data.web.search.searchapi_api_key - app.state.config.SEARCHAPI_ENGINE = form_data.web.search.searchapi_engine + request.app.state.config.SERPSTACK_API_KEY = ( + form_data.web.search.serpstack_api_key + ) + request.app.state.config.SERPSTACK_HTTPS = form_data.web.search.serpstack_https + request.app.state.config.SERPER_API_KEY = form_data.web.search.serper_api_key + request.app.state.config.SERPLY_API_KEY = form_data.web.search.serply_api_key + request.app.state.config.TAVILY_API_KEY = form_data.web.search.tavily_api_key + request.app.state.config.SEARCHAPI_API_KEY = ( + form_data.web.search.searchapi_api_key + ) + request.app.state.config.SEARCHAPI_ENGINE = ( + form_data.web.search.searchapi_engine + ) - app.state.config.JINA_API_KEY = form_data.web.search.jina_api_key - app.state.config.BING_SEARCH_V7_ENDPOINT = ( + request.app.state.config.JINA_API_KEY = form_data.web.search.jina_api_key + request.app.state.config.BING_SEARCH_V7_ENDPOINT = ( form_data.web.search.bing_search_v7_endpoint ) - app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY = ( + request.app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY = ( form_data.web.search.bing_search_v7_subscription_key ) - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT = form_data.web.search.result_count - app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS = ( + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT = ( + form_data.web.search.result_count + ) + request.app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS = ( form_data.web.search.concurrent_requests ) return { "status": True, - "pdf_extract_images": app.state.config.PDF_EXTRACT_IMAGES, + "pdf_extract_images": request.app.state.config.PDF_EXTRACT_IMAGES, "file": { - "max_size": app.state.config.FILE_MAX_SIZE, - "max_count": app.state.config.FILE_MAX_COUNT, + "max_size": request.app.state.config.FILE_MAX_SIZE, + "max_count": request.app.state.config.FILE_MAX_COUNT, }, "content_extraction": { - "engine": app.state.config.CONTENT_EXTRACTION_ENGINE, - "tika_server_url": app.state.config.TIKA_SERVER_URL, + "engine": request.app.state.config.CONTENT_EXTRACTION_ENGINE, + "tika_server_url": request.app.state.config.TIKA_SERVER_URL, }, "chunk": { - "text_splitter": app.state.config.TEXT_SPLITTER, - "chunk_size": app.state.config.CHUNK_SIZE, - "chunk_overlap": app.state.config.CHUNK_OVERLAP, + "text_splitter": request.app.state.config.TEXT_SPLITTER, + "chunk_size": request.app.state.config.CHUNK_SIZE, + "chunk_overlap": request.app.state.config.CHUNK_OVERLAP, }, "youtube": { - "language": app.state.config.YOUTUBE_LOADER_LANGUAGE, - "proxy_url": app.state.config.YOUTUBE_LOADER_PROXY_URL, - "translation": app.state.YOUTUBE_LOADER_TRANSLATION, + "language": request.app.state.config.YOUTUBE_LOADER_LANGUAGE, + "proxy_url": request.app.state.config.YOUTUBE_LOADER_PROXY_URL, + "translation": request.app.state.YOUTUBE_LOADER_TRANSLATION, }, "web": { - "web_loader_ssl_verification": app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION, + "web_loader_ssl_verification": request.app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION, "search": { - "enabled": app.state.config.ENABLE_RAG_WEB_SEARCH, - "engine": app.state.config.RAG_WEB_SEARCH_ENGINE, - "searxng_query_url": app.state.config.SEARXNG_QUERY_URL, - "google_pse_api_key": app.state.config.GOOGLE_PSE_API_KEY, - "google_pse_engine_id": app.state.config.GOOGLE_PSE_ENGINE_ID, - "brave_search_api_key": app.state.config.BRAVE_SEARCH_API_KEY, - "kagi_search_api_key": app.state.config.KAGI_SEARCH_API_KEY, - "mojeek_search_api_key": app.state.config.MOJEEK_SEARCH_API_KEY, - "serpstack_api_key": app.state.config.SERPSTACK_API_KEY, - "serpstack_https": app.state.config.SERPSTACK_HTTPS, - "serper_api_key": app.state.config.SERPER_API_KEY, - "serply_api_key": app.state.config.SERPLY_API_KEY, - "serachapi_api_key": app.state.config.SEARCHAPI_API_KEY, - "searchapi_engine": app.state.config.SEARCHAPI_ENGINE, - "tavily_api_key": app.state.config.TAVILY_API_KEY, - "jina_api_key": app.state.config.JINA_API_KEY, - "bing_search_v7_endpoint": app.state.config.BING_SEARCH_V7_ENDPOINT, - "bing_search_v7_subscription_key": app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY, - "result_count": app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - "concurrent_requests": app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS, + "enabled": request.app.state.config.ENABLE_RAG_WEB_SEARCH, + "engine": request.app.state.config.RAG_WEB_SEARCH_ENGINE, + "searxng_query_url": request.app.state.config.SEARXNG_QUERY_URL, + "google_pse_api_key": request.app.state.config.GOOGLE_PSE_API_KEY, + "google_pse_engine_id": request.app.state.config.GOOGLE_PSE_ENGINE_ID, + "brave_search_api_key": request.app.state.config.BRAVE_SEARCH_API_KEY, + "kagi_search_api_key": request.app.state.config.KAGI_SEARCH_API_KEY, + "mojeek_search_api_key": request.app.state.config.MOJEEK_SEARCH_API_KEY, + "serpstack_api_key": request.app.state.config.SERPSTACK_API_KEY, + "serpstack_https": request.app.state.config.SERPSTACK_HTTPS, + "serper_api_key": request.app.state.config.SERPER_API_KEY, + "serply_api_key": request.app.state.config.SERPLY_API_KEY, + "serachapi_api_key": request.app.state.config.SEARCHAPI_API_KEY, + "searchapi_engine": request.app.state.config.SEARCHAPI_ENGINE, + "tavily_api_key": request.app.state.config.TAVILY_API_KEY, + "jina_api_key": request.app.state.config.JINA_API_KEY, + "bing_search_v7_endpoint": request.app.state.config.BING_SEARCH_V7_ENDPOINT, + "bing_search_v7_subscription_key": request.app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY, + "result_count": request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + "concurrent_requests": request.app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS, }, }, } -@app.get("/template") -async def get_rag_template(user=Depends(get_verified_user)): +@router.get("/template") +async def get_rag_template(request: Request, user=Depends(get_verified_user)): return { "status": True, - "template": app.state.config.RAG_TEMPLATE, + "template": request.app.state.config.RAG_TEMPLATE, } -@app.get("/query/settings") -async def get_query_settings(user=Depends(get_admin_user)): +@router.get("/query/settings") +async def get_query_settings(request: Request, user=Depends(get_admin_user)): return { "status": True, - "template": app.state.config.RAG_TEMPLATE, - "k": app.state.config.TOP_K, - "r": app.state.config.RELEVANCE_THRESHOLD, - "hybrid": app.state.config.ENABLE_RAG_HYBRID_SEARCH, + "template": request.app.state.config.RAG_TEMPLATE, + "k": request.app.state.config.TOP_K, + "r": request.app.state.config.RELEVANCE_THRESHOLD, + "hybrid": request.app.state.config.ENABLE_RAG_HYBRID_SEARCH, } @@ -701,24 +615,24 @@ class QuerySettingsForm(BaseModel): hybrid: Optional[bool] = None -@app.post("/query/settings/update") +@router.post("/query/settings/update") async def update_query_settings( - form_data: QuerySettingsForm, user=Depends(get_admin_user) + request: Request, form_data: QuerySettingsForm, user=Depends(get_admin_user) ): - app.state.config.RAG_TEMPLATE = form_data.template - app.state.config.TOP_K = form_data.k if form_data.k else 4 - app.state.config.RELEVANCE_THRESHOLD = form_data.r if form_data.r else 0.0 + request.app.state.config.RAG_TEMPLATE = form_data.template + request.app.state.config.TOP_K = form_data.k if form_data.k else 4 + request.app.state.config.RELEVANCE_THRESHOLD = form_data.r if form_data.r else 0.0 - app.state.config.ENABLE_RAG_HYBRID_SEARCH = ( + request.app.state.config.ENABLE_RAG_HYBRID_SEARCH = ( form_data.hybrid if form_data.hybrid else False ) return { "status": True, - "template": app.state.config.RAG_TEMPLATE, - "k": app.state.config.TOP_K, - "r": app.state.config.RELEVANCE_THRESHOLD, - "hybrid": app.state.config.ENABLE_RAG_HYBRID_SEARCH, + "template": request.app.state.config.RAG_TEMPLATE, + "k": request.app.state.config.TOP_K, + "r": request.app.state.config.RELEVANCE_THRESHOLD, + "hybrid": request.app.state.config.ENABLE_RAG_HYBRID_SEARCH, } @@ -729,24 +643,8 @@ async def update_query_settings( #################################### -def _get_docs_info(docs: list[Document]) -> str: - docs_info = set() - - # Trying to select relevant metadata identifying the document. - for doc in docs: - metadata = getattr(doc, "metadata", {}) - doc_name = metadata.get("name", "") - if not doc_name: - doc_name = metadata.get("title", "") - if not doc_name: - doc_name = metadata.get("source", "") - if doc_name: - docs_info.add(doc_name) - - return ", ".join(docs_info) - - def save_docs_to_vector_db( + request: Request, docs, collection_name, metadata: Optional[dict] = None, @@ -754,6 +652,22 @@ def save_docs_to_vector_db( split: bool = True, add: bool = False, ) -> bool: + def _get_docs_info(docs: list[Document]) -> str: + docs_info = set() + + # Trying to select relevant metadata identifying the document. + for doc in docs: + metadata = getattr(doc, "metadata", {}) + doc_name = metadata.get("name", "") + if not doc_name: + doc_name = metadata.get("title", "") + if not doc_name: + doc_name = metadata.get("source", "") + if doc_name: + docs_info.add(doc_name) + + return ", ".join(docs_info) + log.info( f"save_docs_to_vector_db: document {_get_docs_info(docs)} {collection_name}" ) @@ -772,22 +686,22 @@ def save_docs_to_vector_db( raise ValueError(ERROR_MESSAGES.DUPLICATE_CONTENT) if split: - if app.state.config.TEXT_SPLITTER in ["", "character"]: + if request.app.state.config.TEXT_SPLITTER in ["", "character"]: text_splitter = RecursiveCharacterTextSplitter( - chunk_size=app.state.config.CHUNK_SIZE, - chunk_overlap=app.state.config.CHUNK_OVERLAP, + chunk_size=request.app.state.config.CHUNK_SIZE, + chunk_overlap=request.app.state.config.CHUNK_OVERLAP, add_start_index=True, ) - elif app.state.config.TEXT_SPLITTER == "token": + elif request.app.state.config.TEXT_SPLITTER == "token": log.info( - f"Using token text splitter: {app.state.config.TIKTOKEN_ENCODING_NAME}" + f"Using token text splitter: {request.app.state.config.TIKTOKEN_ENCODING_NAME}" ) - tiktoken.get_encoding(str(app.state.config.TIKTOKEN_ENCODING_NAME)) + tiktoken.get_encoding(str(request.app.state.config.TIKTOKEN_ENCODING_NAME)) text_splitter = TokenTextSplitter( - encoding_name=str(app.state.config.TIKTOKEN_ENCODING_NAME), - chunk_size=app.state.config.CHUNK_SIZE, - chunk_overlap=app.state.config.CHUNK_OVERLAP, + encoding_name=str(request.app.state.config.TIKTOKEN_ENCODING_NAME), + chunk_size=request.app.state.config.CHUNK_SIZE, + chunk_overlap=request.app.state.config.CHUNK_OVERLAP, add_start_index=True, ) else: @@ -805,8 +719,8 @@ def save_docs_to_vector_db( **(metadata if metadata else {}), "embedding_config": json.dumps( { - "engine": app.state.config.RAG_EMBEDDING_ENGINE, - "model": app.state.config.RAG_EMBEDDING_MODEL, + "engine": request.app.state.config.RAG_EMBEDDING_ENGINE, + "model": request.app.state.config.RAG_EMBEDDING_MODEL, } ), } @@ -835,20 +749,20 @@ def save_docs_to_vector_db( log.info(f"adding to collection {collection_name}") embedding_function = get_embedding_function( - app.state.config.RAG_EMBEDDING_ENGINE, - app.state.config.RAG_EMBEDDING_MODEL, - app.state.sentence_transformer_ef, + request.app.state.config.RAG_EMBEDDING_ENGINE, + request.app.state.config.RAG_EMBEDDING_MODEL, + request.app.state.ef, ( - app.state.config.OPENAI_API_BASE_URL - if app.state.config.RAG_EMBEDDING_ENGINE == "openai" - else app.state.config.OLLAMA_BASE_URL + request.app.state.config.RAG_OPENAI_API_BASE_URL + if request.app.state.config.RAG_EMBEDDING_ENGINE == "openai" + else request.app.state.config.RAG_OLLAMA_BASE_URL ), ( - app.state.config.OPENAI_API_KEY - if app.state.config.RAG_EMBEDDING_ENGINE == "openai" - else app.state.config.OLLAMA_API_KEY + request.app.state.config.RAG_OPENAI_API_KEY + if request.app.state.config.RAG_EMBEDDING_ENGINE == "openai" + else request.app.state.config.RAG_OLLAMA_API_KEY ), - app.state.config.RAG_EMBEDDING_BATCH_SIZE, + request.app.state.config.RAG_EMBEDDING_BATCH_SIZE, ) embeddings = embedding_function( @@ -882,8 +796,9 @@ class ProcessFileForm(BaseModel): collection_name: Optional[str] = None -@app.post("/process/file") +@router.post("/process/file") def process_file( + request: Request, form_data: ProcessFileForm, user=Depends(get_verified_user), ): @@ -953,9 +868,9 @@ def process_file( if file_path: file_path = Storage.get_file(file_path) loader = Loader( - engine=app.state.config.CONTENT_EXTRACTION_ENGINE, - TIKA_SERVER_URL=app.state.config.TIKA_SERVER_URL, - PDF_EXTRACT_IMAGES=app.state.config.PDF_EXTRACT_IMAGES, + engine=request.app.state.config.CONTENT_EXTRACTION_ENGINE, + TIKA_SERVER_URL=request.app.state.config.TIKA_SERVER_URL, + PDF_EXTRACT_IMAGES=request.app.state.config.PDF_EXTRACT_IMAGES, ) docs = loader.load( file.filename, file.meta.get("content_type"), file_path @@ -1000,6 +915,7 @@ def process_file( try: result = save_docs_to_vector_db( + request, docs=docs, collection_name=collection_name, metadata={ @@ -1040,6 +956,479 @@ def process_file( ) +class ProcessTextForm(BaseModel): + name: str + content: str + collection_name: Optional[str] = None + + +@router.post("/process/text") +def process_text( + request: Request, + form_data: ProcessTextForm, + user=Depends(get_verified_user), +): + collection_name = form_data.collection_name + if collection_name is None: + collection_name = calculate_sha256_string(form_data.content) + + docs = [ + Document( + page_content=form_data.content, + metadata={"name": form_data.name, "created_by": user.id}, + ) + ] + text_content = form_data.content + log.debug(f"text_content: {text_content}") + + result = save_docs_to_vector_db(request, docs, collection_name) + if result: + return { + "status": True, + "collection_name": collection_name, + "content": text_content, + } + else: + raise HTTPException( + status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, + detail=ERROR_MESSAGES.DEFAULT(), + ) + + +@router.post("/process/youtube") +def process_youtube_video( + request: Request, form_data: ProcessUrlForm, user=Depends(get_verified_user) +): + try: + collection_name = form_data.collection_name + if not collection_name: + collection_name = calculate_sha256_string(form_data.url)[:63] + + loader = YoutubeLoader( + form_data.url, + language=request.app.state.config.YOUTUBE_LOADER_LANGUAGE, + proxy_url=request.app.state.config.YOUTUBE_LOADER_PROXY_URL, + ) + + docs = loader.load() + content = " ".join([doc.page_content for doc in docs]) + log.debug(f"text_content: {content}") + + save_docs_to_vector_db(request, docs, collection_name, overwrite=True) + + return { + "status": True, + "collection_name": collection_name, + "filename": form_data.url, + "file": { + "data": { + "content": content, + }, + "meta": { + "name": form_data.url, + }, + }, + } + except Exception as e: + log.exception(e) + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=ERROR_MESSAGES.DEFAULT(e), + ) + + +@router.post("/process/web") +def process_web( + request: Request, form_data: ProcessUrlForm, user=Depends(get_verified_user) +): + try: + collection_name = form_data.collection_name + if not collection_name: + collection_name = calculate_sha256_string(form_data.url)[:63] + + loader = get_web_loader( + form_data.url, + verify_ssl=request.app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION, + requests_per_second=request.app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS, + ) + docs = loader.load() + content = " ".join([doc.page_content for doc in docs]) + + log.debug(f"text_content: {content}") + save_docs_to_vector_db(request, docs, collection_name, overwrite=True) + + return { + "status": True, + "collection_name": collection_name, + "filename": form_data.url, + "file": { + "data": { + "content": content, + }, + "meta": { + "name": form_data.url, + }, + }, + } + except Exception as e: + log.exception(e) + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=ERROR_MESSAGES.DEFAULT(e), + ) + + +def search_web(request: Request, engine: str, query: str) -> list[SearchResult]: + """Search the web using a search engine and return the results as a list of SearchResult objects. + Will look for a search engine API key in environment variables in the following order: + - SEARXNG_QUERY_URL + - GOOGLE_PSE_API_KEY + GOOGLE_PSE_ENGINE_ID + - BRAVE_SEARCH_API_KEY + - KAGI_SEARCH_API_KEY + - MOJEEK_SEARCH_API_KEY + - SERPSTACK_API_KEY + - SERPER_API_KEY + - SERPLY_API_KEY + - TAVILY_API_KEY + - SEARCHAPI_API_KEY + SEARCHAPI_ENGINE (by default `google`) + Args: + query (str): The query to search for + """ + + # TODO: add playwright to search the web + if engine == "searxng": + if request.app.state.config.SEARXNG_QUERY_URL: + return search_searxng( + request.app.state.config.SEARXNG_QUERY_URL, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + ) + else: + raise Exception("No SEARXNG_QUERY_URL found in environment variables") + elif engine == "google_pse": + if ( + request.app.state.config.GOOGLE_PSE_API_KEY + and request.app.state.config.GOOGLE_PSE_ENGINE_ID + ): + return search_google_pse( + request.app.state.config.GOOGLE_PSE_API_KEY, + request.app.state.config.GOOGLE_PSE_ENGINE_ID, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + ) + else: + raise Exception( + "No GOOGLE_PSE_API_KEY or GOOGLE_PSE_ENGINE_ID found in environment variables" + ) + elif engine == "brave": + if request.app.state.config.BRAVE_SEARCH_API_KEY: + return search_brave( + request.app.state.config.BRAVE_SEARCH_API_KEY, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + ) + else: + raise Exception("No BRAVE_SEARCH_API_KEY found in environment variables") + elif engine == "kagi": + if request.app.state.config.KAGI_SEARCH_API_KEY: + return search_kagi( + request.app.state.config.KAGI_SEARCH_API_KEY, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + ) + else: + raise Exception("No KAGI_SEARCH_API_KEY found in environment variables") + elif engine == "mojeek": + if request.app.state.config.MOJEEK_SEARCH_API_KEY: + return search_mojeek( + request.app.state.config.MOJEEK_SEARCH_API_KEY, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + ) + else: + raise Exception("No MOJEEK_SEARCH_API_KEY found in environment variables") + elif engine == "serpstack": + if request.app.state.config.SERPSTACK_API_KEY: + return search_serpstack( + request.app.state.config.SERPSTACK_API_KEY, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + https_enabled=request.app.state.config.SERPSTACK_HTTPS, + ) + else: + raise Exception("No SERPSTACK_API_KEY found in environment variables") + elif engine == "serper": + if request.app.state.config.SERPER_API_KEY: + return search_serper( + request.app.state.config.SERPER_API_KEY, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + ) + else: + raise Exception("No SERPER_API_KEY found in environment variables") + elif engine == "serply": + if request.app.state.config.SERPLY_API_KEY: + return search_serply( + request.app.state.config.SERPLY_API_KEY, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + ) + else: + raise Exception("No SERPLY_API_KEY found in environment variables") + elif engine == "duckduckgo": + return search_duckduckgo( + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + ) + elif engine == "tavily": + if request.app.state.config.TAVILY_API_KEY: + return search_tavily( + request.app.state.config.TAVILY_API_KEY, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + ) + else: + raise Exception("No TAVILY_API_KEY found in environment variables") + elif engine == "searchapi": + if request.app.state.config.SEARCHAPI_API_KEY: + return search_searchapi( + request.app.state.config.SEARCHAPI_API_KEY, + request.app.state.config.SEARCHAPI_ENGINE, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + ) + else: + raise Exception("No SEARCHAPI_API_KEY found in environment variables") + elif engine == "jina": + return search_jina( + request.app.state.config.JINA_API_KEY, + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + ) + elif engine == "bing": + return search_bing( + request.app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY, + request.app.state.config.BING_SEARCH_V7_ENDPOINT, + str(DEFAULT_LOCALE), + query, + request.app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, + request.app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, + ) + else: + raise Exception("No search engine API key found in environment variables") + + +@router.post("/process/web/search") +def process_web_search( + request: Request, form_data: SearchForm, user=Depends(get_verified_user) +): + try: + logging.info( + f"trying to web search with {request.app.state.config.RAG_WEB_SEARCH_ENGINE, form_data.query}" + ) + web_results = search_web( + request, request.app.state.config.RAG_WEB_SEARCH_ENGINE, form_data.query + ) + except Exception as e: + log.exception(e) + + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=ERROR_MESSAGES.WEB_SEARCH_ERROR(e), + ) + + try: + collection_name = form_data.collection_name + if collection_name == "": + collection_name = f"web-search-{calculate_sha256_string(form_data.query)}"[ + :63 + ] + + urls = [result.link for result in web_results] + loader = get_web_loader( + urls=urls, + verify_ssl=request.app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION, + requests_per_second=request.app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS, + ) + docs = loader.aload() + + save_docs_to_vector_db(request, docs, collection_name, overwrite=True) + + return { + "status": True, + "collection_name": collection_name, + "filenames": urls, + } + except Exception as e: + log.exception(e) + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=ERROR_MESSAGES.DEFAULT(e), + ) + + +class QueryDocForm(BaseModel): + collection_name: str + query: str + k: Optional[int] = None + r: Optional[float] = None + hybrid: Optional[bool] = None + + +@router.post("/query/doc") +def query_doc_handler( + request: Request, + form_data: QueryDocForm, + user=Depends(get_verified_user), +): + try: + if request.app.state.config.ENABLE_RAG_HYBRID_SEARCH: + return query_doc_with_hybrid_search( + collection_name=form_data.collection_name, + query=form_data.query, + embedding_function=request.app.state.EMBEDDING_FUNCTION, + k=form_data.k if form_data.k else request.app.state.config.TOP_K, + reranking_function=request.app.state.rf, + r=( + form_data.r + if form_data.r + else request.app.state.config.RELEVANCE_THRESHOLD + ), + ) + else: + return query_doc( + collection_name=form_data.collection_name, + query_embedding=request.app.state.EMBEDDING_FUNCTION(form_data.query), + k=form_data.k if form_data.k else request.app.state.config.TOP_K, + ) + except Exception as e: + log.exception(e) + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=ERROR_MESSAGES.DEFAULT(e), + ) + + +class QueryCollectionsForm(BaseModel): + collection_names: list[str] + query: str + k: Optional[int] = None + r: Optional[float] = None + hybrid: Optional[bool] = None + + +@router.post("/query/collection") +def query_collection_handler( + request: Request, + form_data: QueryCollectionsForm, + user=Depends(get_verified_user), +): + try: + if request.app.state.config.ENABLE_RAG_HYBRID_SEARCH: + return query_collection_with_hybrid_search( + collection_names=form_data.collection_names, + queries=[form_data.query], + embedding_function=request.app.state.EMBEDDING_FUNCTION, + k=form_data.k if form_data.k else request.app.state.config.TOP_K, + reranking_function=request.app.state.rf, + r=( + form_data.r + if form_data.r + else request.app.state.config.RELEVANCE_THRESHOLD + ), + ) + else: + return query_collection( + collection_names=form_data.collection_names, + queries=[form_data.query], + embedding_function=request.app.state.EMBEDDING_FUNCTION, + k=form_data.k if form_data.k else request.app.state.config.TOP_K, + ) + + except Exception as e: + log.exception(e) + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=ERROR_MESSAGES.DEFAULT(e), + ) + + +#################################### +# +# Vector DB operations +# +#################################### + + +class DeleteForm(BaseModel): + collection_name: str + file_id: str + + +@router.post("/delete") +def delete_entries_from_collection(form_data: DeleteForm, user=Depends(get_admin_user)): + try: + if VECTOR_DB_CLIENT.has_collection(collection_name=form_data.collection_name): + file = Files.get_file_by_id(form_data.file_id) + hash = file.hash + + VECTOR_DB_CLIENT.delete( + collection_name=form_data.collection_name, + metadata={"hash": hash}, + ) + return {"status": True} + else: + return {"status": False} + except Exception as e: + log.exception(e) + return {"status": False} + + +@router.post("/reset/db") +def reset_vector_db(user=Depends(get_admin_user)): + VECTOR_DB_CLIENT.reset() + Knowledges.delete_all_knowledge() + + +@router.post("/reset/uploads") +def reset_upload_dir(user=Depends(get_admin_user)) -> bool: + folder = f"{UPLOAD_DIR}" + try: + # Check if the directory exists + if os.path.exists(folder): + # Iterate over all the files and directories in the specified directory + for filename in os.listdir(folder): + file_path = os.path.join(folder, filename) + try: + if os.path.isfile(file_path) or os.path.islink(file_path): + os.unlink(file_path) # Remove the file or link + elif os.path.isdir(file_path): + shutil.rmtree(file_path) # Remove the directory + except Exception as e: + print(f"Failed to delete {file_path}. Reason: {e}") + else: + print(f"The directory {folder} does not exist") + except Exception as e: + print(f"Failed to process the directory {folder}. Reason: {e}") + return True + + +if ENV == "dev": + + @router.get("/ef/{text}") + async def get_embeddings(request: Request, text: Optional[str] = "Hello World!"): + return {"result": request.app.state.EMBEDDING_FUNCTION(text)} + class BatchProcessFilesForm(BaseModel): files: List[FileModel] collection_name: str @@ -1053,7 +1442,7 @@ class BatchProcessFilesResponse(BaseModel): results: List[BatchProcessFilesResult] errors: List[BatchProcessFilesResult] -@app.post("/process/files/batch") +@router.post("/process/files/batch") def process_files_batch( form_data: BatchProcessFilesForm, user=Depends(get_verified_user), @@ -1133,466 +1522,3 @@ def process_files_batch( errors=errors ) -class ProcessTextForm(BaseModel): - name: str - content: str - collection_name: Optional[str] = None - - -@app.post("/process/text") -def process_text( - form_data: ProcessTextForm, - user=Depends(get_verified_user), -): - collection_name = form_data.collection_name - if collection_name is None: - collection_name = calculate_sha256_string(form_data.content) - - docs = [ - Document( - page_content=form_data.content, - metadata={"name": form_data.name, "created_by": user.id}, - ) - ] - text_content = form_data.content - log.debug(f"text_content: {text_content}") - - result = save_docs_to_vector_db(docs, collection_name) - - if result: - return { - "status": True, - "collection_name": collection_name, - "content": text_content, - } - else: - raise HTTPException( - status_code=status.HTTP_500_INTERNAL_SERVER_ERROR, - detail=ERROR_MESSAGES.DEFAULT(), - ) - - -@app.post("/process/youtube") -def process_youtube_video(form_data: ProcessUrlForm, user=Depends(get_verified_user)): - try: - collection_name = form_data.collection_name - if not collection_name: - collection_name = calculate_sha256_string(form_data.url)[:63] - - loader = YoutubeLoader( - form_data.url, - language=app.state.config.YOUTUBE_LOADER_LANGUAGE, - proxy_url=app.state.config.YOUTUBE_LOADER_PROXY_URL, - ) - - docs = loader.load() - content = " ".join([doc.page_content for doc in docs]) - log.debug(f"text_content: {content}") - save_docs_to_vector_db(docs, collection_name, overwrite=True) - - return { - "status": True, - "collection_name": collection_name, - "filename": form_data.url, - "file": { - "data": { - "content": content, - }, - "meta": { - "name": form_data.url, - }, - }, - } - except Exception as e: - log.exception(e) - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=ERROR_MESSAGES.DEFAULT(e), - ) - - -@app.post("/process/web") -def process_web(form_data: ProcessUrlForm, user=Depends(get_verified_user)): - try: - collection_name = form_data.collection_name - if not collection_name: - collection_name = calculate_sha256_string(form_data.url)[:63] - - loader = get_web_loader( - form_data.url, - verify_ssl=app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION, - requests_per_second=app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS, - ) - docs = loader.load() - content = " ".join([doc.page_content for doc in docs]) - log.debug(f"text_content: {content}") - save_docs_to_vector_db(docs, collection_name, overwrite=True) - - return { - "status": True, - "collection_name": collection_name, - "filename": form_data.url, - "file": { - "data": { - "content": content, - }, - "meta": { - "name": form_data.url, - }, - }, - } - except Exception as e: - log.exception(e) - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=ERROR_MESSAGES.DEFAULT(e), - ) - - -def search_web(engine: str, query: str) -> list[SearchResult]: - """Search the web using a search engine and return the results as a list of SearchResult objects. - Will look for a search engine API key in environment variables in the following order: - - SEARXNG_QUERY_URL - - GOOGLE_PSE_API_KEY + GOOGLE_PSE_ENGINE_ID - - BRAVE_SEARCH_API_KEY - - KAGI_SEARCH_API_KEY - - MOJEEK_SEARCH_API_KEY - - SERPSTACK_API_KEY - - SERPER_API_KEY - - SERPLY_API_KEY - - TAVILY_API_KEY - - SEARCHAPI_API_KEY + SEARCHAPI_ENGINE (by default `google`) - Args: - query (str): The query to search for - """ - - # TODO: add playwright to search the web - if engine == "searxng": - if app.state.config.SEARXNG_QUERY_URL: - return search_searxng( - app.state.config.SEARXNG_QUERY_URL, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - ) - else: - raise Exception("No SEARXNG_QUERY_URL found in environment variables") - elif engine == "google_pse": - if ( - app.state.config.GOOGLE_PSE_API_KEY - and app.state.config.GOOGLE_PSE_ENGINE_ID - ): - return search_google_pse( - app.state.config.GOOGLE_PSE_API_KEY, - app.state.config.GOOGLE_PSE_ENGINE_ID, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - ) - else: - raise Exception( - "No GOOGLE_PSE_API_KEY or GOOGLE_PSE_ENGINE_ID found in environment variables" - ) - elif engine == "brave": - if app.state.config.BRAVE_SEARCH_API_KEY: - return search_brave( - app.state.config.BRAVE_SEARCH_API_KEY, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - ) - else: - raise Exception("No BRAVE_SEARCH_API_KEY found in environment variables") - elif engine == "kagi": - if app.state.config.KAGI_SEARCH_API_KEY: - return search_kagi( - app.state.config.KAGI_SEARCH_API_KEY, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - ) - else: - raise Exception("No KAGI_SEARCH_API_KEY found in environment variables") - elif engine == "mojeek": - if app.state.config.MOJEEK_SEARCH_API_KEY: - return search_mojeek( - app.state.config.MOJEEK_SEARCH_API_KEY, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - ) - else: - raise Exception("No MOJEEK_SEARCH_API_KEY found in environment variables") - elif engine == "serpstack": - if app.state.config.SERPSTACK_API_KEY: - return search_serpstack( - app.state.config.SERPSTACK_API_KEY, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - https_enabled=app.state.config.SERPSTACK_HTTPS, - ) - else: - raise Exception("No SERPSTACK_API_KEY found in environment variables") - elif engine == "serper": - if app.state.config.SERPER_API_KEY: - return search_serper( - app.state.config.SERPER_API_KEY, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - ) - else: - raise Exception("No SERPER_API_KEY found in environment variables") - elif engine == "serply": - if app.state.config.SERPLY_API_KEY: - return search_serply( - app.state.config.SERPLY_API_KEY, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - ) - else: - raise Exception("No SERPLY_API_KEY found in environment variables") - elif engine == "duckduckgo": - return search_duckduckgo( - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - ) - elif engine == "tavily": - if app.state.config.TAVILY_API_KEY: - return search_tavily( - app.state.config.TAVILY_API_KEY, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - ) - else: - raise Exception("No TAVILY_API_KEY found in environment variables") - elif engine == "searchapi": - if app.state.config.SEARCHAPI_API_KEY: - return search_searchapi( - app.state.config.SEARCHAPI_API_KEY, - app.state.config.SEARCHAPI_ENGINE, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - ) - else: - raise Exception("No SEARCHAPI_API_KEY found in environment variables") - elif engine == "jina": - return search_jina( - app.state.config.JINA_API_KEY, - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - ) - elif engine == "bing": - return search_bing( - app.state.config.BING_SEARCH_V7_SUBSCRIPTION_KEY, - app.state.config.BING_SEARCH_V7_ENDPOINT, - str(DEFAULT_LOCALE), - query, - app.state.config.RAG_WEB_SEARCH_RESULT_COUNT, - app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST, - ) - else: - raise Exception("No search engine API key found in environment variables") - - -@app.post("/process/web/search") -def process_web_search(form_data: SearchForm, user=Depends(get_verified_user)): - try: - logging.info( - f"trying to web search with {app.state.config.RAG_WEB_SEARCH_ENGINE, form_data.query}" - ) - web_results = search_web( - app.state.config.RAG_WEB_SEARCH_ENGINE, form_data.query - ) - except Exception as e: - log.exception(e) - - print(e) - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=ERROR_MESSAGES.WEB_SEARCH_ERROR(e), - ) - - try: - collection_name = form_data.collection_name - if collection_name == "": - collection_name = calculate_sha256_string(form_data.query)[:63] - - urls = [result.link for result in web_results] - - loader = get_web_loader( - urls, - verify_ssl=app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION, - requests_per_second=app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS, - ) - docs = loader.aload() - - save_docs_to_vector_db(docs, collection_name, overwrite=True) - - return { - "status": True, - "collection_name": collection_name, - "filenames": urls, - } - except Exception as e: - log.exception(e) - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=ERROR_MESSAGES.DEFAULT(e), - ) - - -class QueryDocForm(BaseModel): - collection_name: str - query: str - k: Optional[int] = None - r: Optional[float] = None - hybrid: Optional[bool] = None - - -@app.post("/query/doc") -def query_doc_handler( - form_data: QueryDocForm, - user=Depends(get_verified_user), -): - try: - if app.state.config.ENABLE_RAG_HYBRID_SEARCH: - return query_doc_with_hybrid_search( - collection_name=form_data.collection_name, - query=form_data.query, - embedding_function=app.state.EMBEDDING_FUNCTION, - k=form_data.k if form_data.k else app.state.config.TOP_K, - reranking_function=app.state.sentence_transformer_rf, - r=( - form_data.r if form_data.r else app.state.config.RELEVANCE_THRESHOLD - ), - ) - else: - return query_doc( - collection_name=form_data.collection_name, - query_embedding=app.state.EMBEDDING_FUNCTION(form_data.query), - k=form_data.k if form_data.k else app.state.config.TOP_K, - ) - except Exception as e: - log.exception(e) - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=ERROR_MESSAGES.DEFAULT(e), - ) - - -class QueryCollectionsForm(BaseModel): - collection_names: list[str] - query: str - k: Optional[int] = None - r: Optional[float] = None - hybrid: Optional[bool] = None - - -@app.post("/query/collection") -def query_collection_handler( - form_data: QueryCollectionsForm, - user=Depends(get_verified_user), -): - try: - if app.state.config.ENABLE_RAG_HYBRID_SEARCH: - return query_collection_with_hybrid_search( - collection_names=form_data.collection_names, - queries=[form_data.query], - embedding_function=app.state.EMBEDDING_FUNCTION, - k=form_data.k if form_data.k else app.state.config.TOP_K, - reranking_function=app.state.sentence_transformer_rf, - r=( - form_data.r if form_data.r else app.state.config.RELEVANCE_THRESHOLD - ), - ) - else: - return query_collection( - collection_names=form_data.collection_names, - queries=[form_data.query], - embedding_function=app.state.EMBEDDING_FUNCTION, - k=form_data.k if form_data.k else app.state.config.TOP_K, - ) - - except Exception as e: - log.exception(e) - raise HTTPException( - status_code=status.HTTP_400_BAD_REQUEST, - detail=ERROR_MESSAGES.DEFAULT(e), - ) - - -#################################### -# -# Vector DB operations -# -#################################### - - -class DeleteForm(BaseModel): - collection_name: str - file_id: str - - -@app.post("/delete") -def delete_entries_from_collection(form_data: DeleteForm, user=Depends(get_admin_user)): - try: - if VECTOR_DB_CLIENT.has_collection(collection_name=form_data.collection_name): - file = Files.get_file_by_id(form_data.file_id) - hash = file.hash - - VECTOR_DB_CLIENT.delete( - collection_name=form_data.collection_name, - metadata={"hash": hash}, - ) - return {"status": True} - else: - return {"status": False} - except Exception as e: - log.exception(e) - return {"status": False} - - -@app.post("/reset/db") -def reset_vector_db(user=Depends(get_admin_user)): - VECTOR_DB_CLIENT.reset() - Knowledges.delete_all_knowledge() - - -@app.post("/reset/uploads") -def reset_upload_dir(user=Depends(get_admin_user)) -> bool: - folder = f"{UPLOAD_DIR}" - try: - # Check if the directory exists - if os.path.exists(folder): - # Iterate over all the files and directories in the specified directory - for filename in os.listdir(folder): - file_path = os.path.join(folder, filename) - try: - if os.path.isfile(file_path) or os.path.islink(file_path): - os.unlink(file_path) # Remove the file or link - elif os.path.isdir(file_path): - shutil.rmtree(file_path) # Remove the directory - except Exception as e: - print(f"Failed to delete {file_path}. Reason: {e}") - else: - print(f"The directory {folder} does not exist") - except Exception as e: - print(f"Failed to process the directory {folder}. Reason: {e}") - return True - - -if ENV == "dev": - - @app.get("/ef") - async def get_embeddings(): - return {"result": app.state.EMBEDDING_FUNCTION("hello world")} - - @app.get("/ef/{text}") - async def get_embeddings_text(text: str): - return {"result": app.state.EMBEDDING_FUNCTION(text)} - diff --git a/backend/open_webui/routers/tasks.py b/backend/open_webui/routers/tasks.py new file mode 100644 index 000000000..a2a6cdc92 --- /dev/null +++ b/backend/open_webui/routers/tasks.py @@ -0,0 +1,512 @@ +from fastapi import APIRouter, Depends, HTTPException, Response, status, Request +from fastapi.responses import JSONResponse, RedirectResponse + +from pydantic import BaseModel +from typing import Optional +import logging + +from open_webui.utils.chat import generate_chat_completion +from open_webui.utils.task import ( + title_generation_template, + query_generation_template, + autocomplete_generation_template, + tags_generation_template, + emoji_generation_template, + moa_response_generation_template, +) +from open_webui.utils.auth import get_admin_user, get_verified_user +from open_webui.constants import TASKS + +from open_webui.routers.pipelines import process_pipeline_inlet_filter +from open_webui.utils.task import get_task_model_id + +from open_webui.config import ( + DEFAULT_TITLE_GENERATION_PROMPT_TEMPLATE, + DEFAULT_TAGS_GENERATION_PROMPT_TEMPLATE, + DEFAULT_QUERY_GENERATION_PROMPT_TEMPLATE, + DEFAULT_AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE, + DEFAULT_EMOJI_GENERATION_PROMPT_TEMPLATE, + DEFAULT_MOA_GENERATION_PROMPT_TEMPLATE, +) +from open_webui.env import SRC_LOG_LEVELS + + +log = logging.getLogger(__name__) +log.setLevel(SRC_LOG_LEVELS["MODELS"]) + +router = APIRouter() + + +################################## +# +# Task Endpoints +# +################################## + + +@router.get("/config") +async def get_task_config(request: Request, user=Depends(get_verified_user)): + return { + "TASK_MODEL": request.app.state.config.TASK_MODEL, + "TASK_MODEL_EXTERNAL": request.app.state.config.TASK_MODEL_EXTERNAL, + "TITLE_GENERATION_PROMPT_TEMPLATE": request.app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE, + "ENABLE_AUTOCOMPLETE_GENERATION": request.app.state.config.ENABLE_AUTOCOMPLETE_GENERATION, + "AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH": request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH, + "TAGS_GENERATION_PROMPT_TEMPLATE": request.app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE, + "ENABLE_TAGS_GENERATION": request.app.state.config.ENABLE_TAGS_GENERATION, + "ENABLE_SEARCH_QUERY_GENERATION": request.app.state.config.ENABLE_SEARCH_QUERY_GENERATION, + "ENABLE_RETRIEVAL_QUERY_GENERATION": request.app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION, + "QUERY_GENERATION_PROMPT_TEMPLATE": request.app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE, + "TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE": request.app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE, + } + + +class TaskConfigForm(BaseModel): + TASK_MODEL: Optional[str] + TASK_MODEL_EXTERNAL: Optional[str] + TITLE_GENERATION_PROMPT_TEMPLATE: str + ENABLE_AUTOCOMPLETE_GENERATION: bool + AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH: int + TAGS_GENERATION_PROMPT_TEMPLATE: str + ENABLE_TAGS_GENERATION: bool + ENABLE_SEARCH_QUERY_GENERATION: bool + ENABLE_RETRIEVAL_QUERY_GENERATION: bool + QUERY_GENERATION_PROMPT_TEMPLATE: str + TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE: str + + +@router.post("/config/update") +async def update_task_config( + request: Request, form_data: TaskConfigForm, user=Depends(get_admin_user) +): + request.app.state.config.TASK_MODEL = form_data.TASK_MODEL + request.app.state.config.TASK_MODEL_EXTERNAL = form_data.TASK_MODEL_EXTERNAL + request.app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE = ( + form_data.TITLE_GENERATION_PROMPT_TEMPLATE + ) + + request.app.state.config.ENABLE_AUTOCOMPLETE_GENERATION = ( + form_data.ENABLE_AUTOCOMPLETE_GENERATION + ) + request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH = ( + form_data.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH + ) + + request.app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE = ( + form_data.TAGS_GENERATION_PROMPT_TEMPLATE + ) + request.app.state.config.ENABLE_TAGS_GENERATION = form_data.ENABLE_TAGS_GENERATION + request.app.state.config.ENABLE_SEARCH_QUERY_GENERATION = ( + form_data.ENABLE_SEARCH_QUERY_GENERATION + ) + request.app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION = ( + form_data.ENABLE_RETRIEVAL_QUERY_GENERATION + ) + + request.app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE = ( + form_data.QUERY_GENERATION_PROMPT_TEMPLATE + ) + request.app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE = ( + form_data.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE + ) + + return { + "TASK_MODEL": request.app.state.config.TASK_MODEL, + "TASK_MODEL_EXTERNAL": request.app.state.config.TASK_MODEL_EXTERNAL, + "TITLE_GENERATION_PROMPT_TEMPLATE": request.app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE, + "ENABLE_AUTOCOMPLETE_GENERATION": request.app.state.config.ENABLE_AUTOCOMPLETE_GENERATION, + "AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH": request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH, + "TAGS_GENERATION_PROMPT_TEMPLATE": request.app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE, + "ENABLE_TAGS_GENERATION": request.app.state.config.ENABLE_TAGS_GENERATION, + "ENABLE_SEARCH_QUERY_GENERATION": request.app.state.config.ENABLE_SEARCH_QUERY_GENERATION, + "ENABLE_RETRIEVAL_QUERY_GENERATION": request.app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION, + "QUERY_GENERATION_PROMPT_TEMPLATE": request.app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE, + "TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE": request.app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE, + } + + +@router.post("/title/completions") +async def generate_title( + request: Request, form_data: dict, user=Depends(get_verified_user) +): + models = request.app.state.MODELS + + model_id = form_data["model"] + if model_id not in models: + raise HTTPException( + status_code=status.HTTP_404_NOT_FOUND, + detail="Model not found", + ) + + # Check if the user has a custom task model + # If the user has a custom task model, use that model + task_model_id = get_task_model_id( + model_id, + request.app.state.config.TASK_MODEL, + request.app.state.config.TASK_MODEL_EXTERNAL, + models, + ) + + log.debug( + f"generating chat title using model {task_model_id} for user {user.email} " + ) + + if request.app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE != "": + template = request.app.state.config.TITLE_GENERATION_PROMPT_TEMPLATE + else: + template = DEFAULT_TITLE_GENERATION_PROMPT_TEMPLATE + + content = title_generation_template( + template, + form_data["messages"], + { + "name": user.name, + "location": user.info.get("location") if user.info else None, + }, + ) + + payload = { + "model": task_model_id, + "messages": [{"role": "user", "content": content}], + "stream": False, + **( + {"max_tokens": 50} + if models[task_model_id]["owned_by"] == "ollama" + else { + "max_completion_tokens": 50, + } + ), + "metadata": { + "task": str(TASKS.TITLE_GENERATION), + "task_body": form_data, + "chat_id": form_data.get("chat_id", None), + }, + } + + try: + return await generate_chat_completion(request, form_data=payload, user=user) + except Exception as e: + return JSONResponse( + status_code=status.HTTP_400_BAD_REQUEST, + content={"detail": str(e)}, + ) + + +@router.post("/tags/completions") +async def generate_chat_tags( + request: Request, form_data: dict, user=Depends(get_verified_user) +): + + if not request.app.state.config.ENABLE_TAGS_GENERATION: + return JSONResponse( + status_code=status.HTTP_200_OK, + content={"detail": "Tags generation is disabled"}, + ) + + models = request.app.state.MODELS + + model_id = form_data["model"] + if model_id not in models: + raise HTTPException( + status_code=status.HTTP_404_NOT_FOUND, + detail="Model not found", + ) + + # Check if the user has a custom task model + # If the user has a custom task model, use that model + task_model_id = get_task_model_id( + model_id, + request.app.state.config.TASK_MODEL, + request.app.state.config.TASK_MODEL_EXTERNAL, + models, + ) + + log.debug( + f"generating chat tags using model {task_model_id} for user {user.email} " + ) + + if request.app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE != "": + template = request.app.state.config.TAGS_GENERATION_PROMPT_TEMPLATE + else: + template = DEFAULT_TAGS_GENERATION_PROMPT_TEMPLATE + + content = tags_generation_template( + template, form_data["messages"], {"name": user.name} + ) + + payload = { + "model": task_model_id, + "messages": [{"role": "user", "content": content}], + "stream": False, + "metadata": { + "task": str(TASKS.TAGS_GENERATION), + "task_body": form_data, + "chat_id": form_data.get("chat_id", None), + }, + } + + try: + return await generate_chat_completion(request, form_data=payload, user=user) + except Exception as e: + return JSONResponse( + status_code=status.HTTP_400_BAD_REQUEST, + content={"detail": str(e)}, + ) + + +@router.post("/queries/completions") +async def generate_queries( + request: Request, form_data: dict, user=Depends(get_verified_user) +): + + type = form_data.get("type") + if type == "web_search": + if not request.app.state.config.ENABLE_SEARCH_QUERY_GENERATION: + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=f"Search query generation is disabled", + ) + elif type == "retrieval": + if not request.app.state.config.ENABLE_RETRIEVAL_QUERY_GENERATION: + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=f"Query generation is disabled", + ) + + models = request.app.state.MODELS + + model_id = form_data["model"] + if model_id not in models: + raise HTTPException( + status_code=status.HTTP_404_NOT_FOUND, + detail="Model not found", + ) + + # Check if the user has a custom task model + # If the user has a custom task model, use that model + task_model_id = get_task_model_id( + model_id, + request.app.state.config.TASK_MODEL, + request.app.state.config.TASK_MODEL_EXTERNAL, + models, + ) + + log.debug( + f"generating {type} queries using model {task_model_id} for user {user.email}" + ) + + if (request.app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE).strip() != "": + template = request.app.state.config.QUERY_GENERATION_PROMPT_TEMPLATE + else: + template = DEFAULT_QUERY_GENERATION_PROMPT_TEMPLATE + + content = query_generation_template( + template, form_data["messages"], {"name": user.name} + ) + + payload = { + "model": task_model_id, + "messages": [{"role": "user", "content": content}], + "stream": False, + "metadata": { + "task": str(TASKS.QUERY_GENERATION), + "task_body": form_data, + "chat_id": form_data.get("chat_id", None), + }, + } + + try: + return await generate_chat_completion(request, form_data=payload, user=user) + except Exception as e: + return JSONResponse( + status_code=status.HTTP_400_BAD_REQUEST, + content={"detail": str(e)}, + ) + + +@router.post("/auto/completions") +async def generate_autocompletion( + request: Request, form_data: dict, user=Depends(get_verified_user) +): + if not request.app.state.config.ENABLE_AUTOCOMPLETE_GENERATION: + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=f"Autocompletion generation is disabled", + ) + + type = form_data.get("type") + prompt = form_data.get("prompt") + messages = form_data.get("messages") + + if request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH > 0: + if ( + len(prompt) + > request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH + ): + raise HTTPException( + status_code=status.HTTP_400_BAD_REQUEST, + detail=f"Input prompt exceeds maximum length of {request.app.state.config.AUTOCOMPLETE_GENERATION_INPUT_MAX_LENGTH}", + ) + + models = request.app.state.MODELS + + model_id = form_data["model"] + if model_id not in models: + raise HTTPException( + status_code=status.HTTP_404_NOT_FOUND, + detail="Model not found", + ) + + # Check if the user has a custom task model + # If the user has a custom task model, use that model + task_model_id = get_task_model_id( + model_id, + request.app.state.config.TASK_MODEL, + request.app.state.config.TASK_MODEL_EXTERNAL, + models, + ) + + log.debug( + f"generating autocompletion using model {task_model_id} for user {user.email}" + ) + + if (request.app.state.config.AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE).strip() != "": + template = request.app.state.config.AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE + else: + template = DEFAULT_AUTOCOMPLETE_GENERATION_PROMPT_TEMPLATE + + content = autocomplete_generation_template( + template, prompt, messages, type, {"name": user.name} + ) + + payload = { + "model": task_model_id, + "messages": [{"role": "user", "content": content}], + "stream": False, + "metadata": { + "task": str(TASKS.AUTOCOMPLETE_GENERATION), + "task_body": form_data, + "chat_id": form_data.get("chat_id", None), + }, + } + + try: + return await generate_chat_completion(request, form_data=payload, user=user) + except Exception as e: + return JSONResponse( + status_code=status.HTTP_400_BAD_REQUEST, + content={"detail": str(e)}, + ) + + +@router.post("/emoji/completions") +async def generate_emoji( + request: Request, form_data: dict, user=Depends(get_verified_user) +): + + models = request.app.state.MODELS + + model_id = form_data["model"] + if model_id not in models: + raise HTTPException( + status_code=status.HTTP_404_NOT_FOUND, + detail="Model not found", + ) + + # Check if the user has a custom task model + # If the user has a custom task model, use that model + task_model_id = get_task_model_id( + model_id, + request.app.state.config.TASK_MODEL, + request.app.state.config.TASK_MODEL_EXTERNAL, + models, + ) + + log.debug(f"generating emoji using model {task_model_id} for user {user.email} ") + + template = DEFAULT_EMOJI_GENERATION_PROMPT_TEMPLATE + + content = emoji_generation_template( + template, + form_data["prompt"], + { + "name": user.name, + "location": user.info.get("location") if user.info else None, + }, + ) + + payload = { + "model": task_model_id, + "messages": [{"role": "user", "content": content}], + "stream": False, + **( + {"max_tokens": 4} + if models[task_model_id]["owned_by"] == "ollama" + else { + "max_completion_tokens": 4, + } + ), + "chat_id": form_data.get("chat_id", None), + "metadata": {"task": str(TASKS.EMOJI_GENERATION), "task_body": form_data}, + } + + try: + return await generate_chat_completion(request, form_data=payload, user=user) + except Exception as e: + return JSONResponse( + status_code=status.HTTP_400_BAD_REQUEST, + content={"detail": str(e)}, + ) + + +@router.post("/moa/completions") +async def generate_moa_response( + request: Request, form_data: dict, user=Depends(get_verified_user) +): + + models = request.app.state.MODELS + model_id = form_data["model"] + + if model_id not in models: + raise HTTPException( + status_code=status.HTTP_404_NOT_FOUND, + detail="Model not found", + ) + + # Check if the user has a custom task model + # If the user has a custom task model, use that model + task_model_id = get_task_model_id( + model_id, + request.app.state.config.TASK_MODEL, + request.app.state.config.TASK_MODEL_EXTERNAL, + models, + ) + + log.debug(f"generating MOA model {task_model_id} for user {user.email} ") + + template = DEFAULT_MOA_GENERATION_PROMPT_TEMPLATE + + content = moa_response_generation_template( + template, + form_data["prompt"], + form_data["responses"], + ) + + payload = { + "model": task_model_id, + "messages": [{"role": "user", "content": content}], + "stream": form_data.get("stream", False), + "chat_id": form_data.get("chat_id", None), + "metadata": { + "task": str(TASKS.MOA_RESPONSE_GENERATION), + "task_body": form_data, + }, + } + + try: + return await generate_chat_completion(request, form_data=payload, user=user) + except Exception as e: + return JSONResponse( + status_code=status.HTTP_400_BAD_REQUEST, + content={"detail": str(e)}, + ) diff --git a/backend/open_webui/apps/webui/routers/tools.py b/backend/open_webui/routers/tools.py similarity index 98% rename from backend/open_webui/apps/webui/routers/tools.py rename to backend/open_webui/routers/tools.py index 410f12d64..9e95ebe5a 100644 --- a/backend/open_webui/apps/webui/routers/tools.py +++ b/backend/open_webui/routers/tools.py @@ -1,14 +1,14 @@ from pathlib import Path from typing import Optional -from open_webui.apps.webui.models.tools import ( +from open_webui.models.tools import ( ToolForm, ToolModel, ToolResponse, ToolUserResponse, Tools, ) -from open_webui.apps.webui.utils import load_tools_module_by_id, replace_imports +from open_webui.utils.plugin import load_tools_module_by_id, replace_imports from open_webui.config import CACHE_DIR from open_webui.constants import ERROR_MESSAGES from fastapi import APIRouter, Depends, HTTPException, Request, status diff --git a/backend/open_webui/apps/webui/routers/users.py b/backend/open_webui/routers/users.py similarity index 98% rename from backend/open_webui/apps/webui/routers/users.py rename to backend/open_webui/routers/users.py index 92131b9ad..1206d56f2 100644 --- a/backend/open_webui/apps/webui/routers/users.py +++ b/backend/open_webui/routers/users.py @@ -1,9 +1,9 @@ import logging from typing import Optional -from open_webui.apps.webui.models.auths import Auths -from open_webui.apps.webui.models.chats import Chats -from open_webui.apps.webui.models.users import ( +from open_webui.models.auths import Auths +from open_webui.models.chats import Chats +from open_webui.models.users import ( UserModel, UserRoleUpdateForm, Users, diff --git a/backend/open_webui/apps/webui/routers/utils.py b/backend/open_webui/routers/utils.py similarity index 95% rename from backend/open_webui/apps/webui/routers/utils.py rename to backend/open_webui/routers/utils.py index a4c33a03b..ea73e9759 100644 --- a/backend/open_webui/apps/webui/routers/utils.py +++ b/backend/open_webui/routers/utils.py @@ -1,7 +1,7 @@ import black import markdown -from open_webui.apps.webui.models.chats import ChatTitleMessagesForm +from open_webui.models.chats import ChatTitleMessagesForm from open_webui.config import DATA_DIR, ENABLE_ADMIN_EXPORT from open_webui.constants import ERROR_MESSAGES from fastapi import APIRouter, Depends, HTTPException, Response, status @@ -76,7 +76,7 @@ async def download_db(user=Depends(get_admin_user)): status_code=status.HTTP_401_UNAUTHORIZED, detail=ERROR_MESSAGES.ACCESS_PROHIBITED, ) - from open_webui.apps.webui.internal.db import engine + from open_webui.internal.db import engine if engine.name != "sqlite": raise HTTPException( diff --git a/backend/open_webui/apps/socket/main.py b/backend/open_webui/socket/main.py similarity index 97% rename from backend/open_webui/apps/socket/main.py rename to backend/open_webui/socket/main.py index 8ec8937a1..c0f45c9a0 100644 --- a/backend/open_webui/apps/socket/main.py +++ b/backend/open_webui/socket/main.py @@ -6,14 +6,14 @@ import logging import sys import time -from open_webui.apps.webui.models.users import Users +from open_webui.models.users import Users from open_webui.env import ( ENABLE_WEBSOCKET_SUPPORT, WEBSOCKET_MANAGER, WEBSOCKET_REDIS_URL, ) from open_webui.utils.auth import decode_token -from open_webui.apps.socket.utils import RedisDict +from open_webui.socket.utils import RedisDict from open_webui.env import ( GLOBAL_LOG_LEVEL, @@ -173,6 +173,11 @@ async def user_count(sid): await sio.emit("user-count", {"count": len(USER_POOL.items())}) +@sio.on("chat") +async def chat(sid, data): + print("chat", sid, SESSION_POOL[sid], data) + + @sio.event async def disconnect(sid): if sid in SESSION_POOL: diff --git a/backend/open_webui/apps/socket/utils.py b/backend/open_webui/socket/utils.py similarity index 100% rename from backend/open_webui/apps/socket/utils.py rename to backend/open_webui/socket/utils.py diff --git a/backend/open_webui/test/apps/webui/routers/test_auths.py b/backend/open_webui/test/apps/webui/routers/test_auths.py index cee68228e..f0f69e26d 100644 --- a/backend/open_webui/test/apps/webui/routers/test_auths.py +++ b/backend/open_webui/test/apps/webui/routers/test_auths.py @@ -7,8 +7,8 @@ class TestAuths(AbstractPostgresTest): def setup_class(cls): super().setup_class() - from open_webui.apps.webui.models.auths import Auths - from open_webui.apps.webui.models.users import Users + from open_webui.models.auths import Auths + from open_webui.models.users import Users cls.users = Users cls.auths = Auths diff --git a/backend/open_webui/test/apps/webui/routers/test_chats.py b/backend/open_webui/test/apps/webui/routers/test_chats.py index 935316fd8..a36a01fb1 100644 --- a/backend/open_webui/test/apps/webui/routers/test_chats.py +++ b/backend/open_webui/test/apps/webui/routers/test_chats.py @@ -12,7 +12,7 @@ class TestChats(AbstractPostgresTest): def setup_method(self): super().setup_method() - from open_webui.apps.webui.models.chats import ChatForm, Chats + from open_webui.models.chats import ChatForm, Chats self.chats = Chats self.chats.insert_new_chat( @@ -88,7 +88,7 @@ class TestChats(AbstractPostgresTest): def test_get_user_archived_chats(self): self.chats.archive_all_chats_by_user_id("2") - from open_webui.apps.webui.internal.db import Session + from open_webui.internal.db import Session Session.commit() with mock_webui_user(id="2"): diff --git a/backend/open_webui/test/apps/webui/routers/test_models.py b/backend/open_webui/test/apps/webui/routers/test_models.py index 1d52658b8..c16ca9d07 100644 --- a/backend/open_webui/test/apps/webui/routers/test_models.py +++ b/backend/open_webui/test/apps/webui/routers/test_models.py @@ -7,7 +7,7 @@ class TestModels(AbstractPostgresTest): def setup_class(cls): super().setup_class() - from open_webui.apps.webui.models.models import Model + from open_webui.models.models import Model cls.models = Model diff --git a/backend/open_webui/test/apps/webui/routers/test_users.py b/backend/open_webui/test/apps/webui/routers/test_users.py index 6facf7055..1a58ab147 100644 --- a/backend/open_webui/test/apps/webui/routers/test_users.py +++ b/backend/open_webui/test/apps/webui/routers/test_users.py @@ -25,7 +25,7 @@ class TestUsers(AbstractPostgresTest): def setup_class(cls): super().setup_class() - from open_webui.apps.webui.models.users import Users + from open_webui.models.users import Users cls.users = Users diff --git a/backend/open_webui/test/util/abstract_integration_test.py b/backend/open_webui/test/util/abstract_integration_test.py index 2814731e0..e8492befb 100644 --- a/backend/open_webui/test/util/abstract_integration_test.py +++ b/backend/open_webui/test/util/abstract_integration_test.py @@ -115,7 +115,7 @@ class AbstractPostgresTest(AbstractIntegrationTest): pytest.fail(f"Could not setup test environment: {ex}") def _check_db_connection(self): - from open_webui.apps.webui.internal.db import Session + from open_webui.internal.db import Session retries = 10 while retries > 0: @@ -139,7 +139,7 @@ class AbstractPostgresTest(AbstractIntegrationTest): cls.docker_client.containers.get(cls.DOCKER_CONTAINER_NAME).remove(force=True) def teardown_method(self): - from open_webui.apps.webui.internal.db import Session + from open_webui.internal.db import Session # rollback everything not yet committed Session.commit() diff --git a/backend/open_webui/test/util/mock_user.py b/backend/open_webui/test/util/mock_user.py index ba8e24d4e..7ce64dffa 100644 --- a/backend/open_webui/test/util/mock_user.py +++ b/backend/open_webui/test/util/mock_user.py @@ -5,7 +5,7 @@ from fastapi import FastAPI @contextmanager def mock_webui_user(**kwargs): - from open_webui.apps.webui.main import app + from open_webui.routers.webui import app with mock_user(app, **kwargs): yield @@ -19,7 +19,7 @@ def mock_user(app: FastAPI, **kwargs): get_admin_user, get_current_user_by_api_key, ) - from open_webui.apps.webui.models.users import User + from open_webui.models.users import User def create_user(): user_parameters = { diff --git a/backend/open_webui/utils/access_control.py b/backend/open_webui/utils/access_control.py index 270b28bcc..3b3e75a8b 100644 --- a/backend/open_webui/utils/access_control.py +++ b/backend/open_webui/utils/access_control.py @@ -1,5 +1,5 @@ from typing import Optional, Union, List, Dict, Any -from open_webui.apps.webui.models.groups import Groups +from open_webui.models.groups import Groups import json diff --git a/backend/open_webui/utils/auth.py b/backend/open_webui/utils/auth.py index cde953102..e1a0ca671 100644 --- a/backend/open_webui/utils/auth.py +++ b/backend/open_webui/utils/auth.py @@ -5,7 +5,7 @@ import jwt from datetime import UTC, datetime, timedelta from typing import Optional, Union, List, Dict -from open_webui.apps.webui.models.users import Users +from open_webui.models.users import Users from open_webui.constants import ERROR_MESSAGES from open_webui.env import WEBUI_SECRET_KEY diff --git a/backend/open_webui/utils/chat.py b/backend/open_webui/utils/chat.py new file mode 100644 index 000000000..56904d1d8 --- /dev/null +++ b/backend/open_webui/utils/chat.py @@ -0,0 +1,374 @@ +import time +import logging +import sys + +from aiocache import cached +from typing import Any, Optional +import random +import json +import inspect + +from fastapi import Request +from starlette.responses import Response, StreamingResponse + + +from open_webui.models.users import UserModel + +from open_webui.socket.main import ( + get_event_call, + get_event_emitter, +) +from open_webui.functions import generate_function_chat_completion + +from open_webui.routers.openai import ( + generate_chat_completion as generate_openai_chat_completion, +) + +from open_webui.routers.ollama import ( + generate_chat_completion as generate_ollama_chat_completion, +) + +from open_webui.routers.pipelines import ( + process_pipeline_inlet_filter, + process_pipeline_outlet_filter, +) + +from open_webui.models.functions import Functions +from open_webui.models.models import Models + + +from open_webui.utils.plugin import load_function_module_by_id +from open_webui.utils.models import get_all_models, check_model_access +from open_webui.utils.payload import convert_payload_openai_to_ollama +from open_webui.utils.response import ( + convert_response_ollama_to_openai, + convert_streaming_response_ollama_to_openai, +) + +from open_webui.env import SRC_LOG_LEVELS, GLOBAL_LOG_LEVEL, BYPASS_MODEL_ACCESS_CONTROL + + +logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL) +log = logging.getLogger(__name__) +log.setLevel(SRC_LOG_LEVELS["MAIN"]) + + +async def generate_chat_completion( + request: Request, + form_data: dict, + user: Any, + bypass_filter: bool = False, +): + if BYPASS_MODEL_ACCESS_CONTROL: + bypass_filter = True + + models = request.app.state.MODELS + + model_id = form_data["model"] + if model_id not in models: + raise Exception("Model not found") + + # Process the form_data through the pipeline + try: + form_data = process_pipeline_inlet_filter(request, form_data, user, models) + except Exception as e: + raise e + + model = models[model_id] + + # Check if user has access to the model + if not bypass_filter and user.role == "user": + try: + check_model_access(user, model) + except Exception as e: + raise e + + if model["owned_by"] == "arena": + model_ids = model.get("info", {}).get("meta", {}).get("model_ids") + filter_mode = model.get("info", {}).get("meta", {}).get("filter_mode") + if model_ids and filter_mode == "exclude": + model_ids = [ + model["id"] + for model in await get_all_models(request) + if model.get("owned_by") != "arena" and model["id"] not in model_ids + ] + + selected_model_id = None + if isinstance(model_ids, list) and model_ids: + selected_model_id = random.choice(model_ids) + else: + model_ids = [ + model["id"] + for model in await get_all_models(request) + if model.get("owned_by") != "arena" + ] + selected_model_id = random.choice(model_ids) + + form_data["model"] = selected_model_id + + if form_data.get("stream") == True: + + async def stream_wrapper(stream): + yield f"data: {json.dumps({'selected_model_id': selected_model_id})}\n\n" + async for chunk in stream: + yield chunk + + response = await generate_chat_completion( + form_data, user, bypass_filter=True + ) + return StreamingResponse( + stream_wrapper(response.body_iterator), media_type="text/event-stream" + ) + else: + return { + **(await generate_chat_completion(form_data, user, bypass_filter=True)), + "selected_model_id": selected_model_id, + } + + if model.get("pipe"): + # Below does not require bypass_filter because this is the only route the uses this function and it is already bypassing the filter + return await generate_function_chat_completion( + form_data, user=user, models=models + ) + if model["owned_by"] == "ollama": + # Using /ollama/api/chat endpoint + form_data = convert_payload_openai_to_ollama(form_data) + response = await generate_ollama_chat_completion( + request=request, form_data=form_data, user=user, bypass_filter=bypass_filter + ) + if form_data.get("stream"): + response.headers["content-type"] = "text/event-stream" + return StreamingResponse( + convert_streaming_response_ollama_to_openai(response), + headers=dict(response.headers), + ) + else: + return convert_response_ollama_to_openai(response) + else: + return await generate_openai_chat_completion( + request=request, form_data=form_data, user=user, bypass_filter=bypass_filter + ) + + +async def chat_completed(request: Request, form_data: dict, user: Any): + await get_all_models(request) + models = request.app.state.MODELS + + data = form_data + model_id = data["model"] + if model_id not in models: + raise Exception("Model not found") + + model = models[model_id] + + try: + data = process_pipeline_outlet_filter(request, data, user, models) + except Exception as e: + return Exception(f"Error: {e}") + + __event_emitter__ = get_event_emitter( + { + "chat_id": data["chat_id"], + "message_id": data["id"], + "session_id": data["session_id"], + } + ) + + __event_call__ = get_event_call( + { + "chat_id": data["chat_id"], + "message_id": data["id"], + "session_id": data["session_id"], + } + ) + + def get_priority(function_id): + function = Functions.get_function_by_id(function_id) + if function is not None and hasattr(function, "valves"): + # TODO: Fix FunctionModel to include vavles + return (function.valves if function.valves else {}).get("priority", 0) + return 0 + + filter_ids = [function.id for function in Functions.get_global_filter_functions()] + if "info" in model and "meta" in model["info"]: + filter_ids.extend(model["info"]["meta"].get("filterIds", [])) + filter_ids = list(set(filter_ids)) + + enabled_filter_ids = [ + function.id + for function in Functions.get_functions_by_type("filter", active_only=True) + ] + filter_ids = [ + filter_id for filter_id in filter_ids if filter_id in enabled_filter_ids + ] + + # Sort filter_ids by priority, using the get_priority function + filter_ids.sort(key=get_priority) + + for filter_id in filter_ids: + filter = Functions.get_function_by_id(filter_id) + if not filter: + continue + + if filter_id in request.app.state.FUNCTIONS: + function_module = request.app.state.FUNCTIONS[filter_id] + else: + function_module, _, _ = load_function_module_by_id(filter_id) + request.app.state.FUNCTIONS[filter_id] = function_module + + if hasattr(function_module, "valves") and hasattr(function_module, "Valves"): + valves = Functions.get_function_valves_by_id(filter_id) + function_module.valves = function_module.Valves( + **(valves if valves else {}) + ) + + if not hasattr(function_module, "outlet"): + continue + try: + outlet = function_module.outlet + + # Get the signature of the function + sig = inspect.signature(outlet) + params = {"body": data} + + # Extra parameters to be passed to the function + extra_params = { + "__model__": model, + "__id__": filter_id, + "__event_emitter__": __event_emitter__, + "__event_call__": __event_call__, + "__request__": request, + } + + # Add extra params in contained in function signature + for key, value in extra_params.items(): + if key in sig.parameters: + params[key] = value + + if "__user__" in sig.parameters: + __user__ = { + "id": user.id, + "email": user.email, + "name": user.name, + "role": user.role, + } + + try: + if hasattr(function_module, "UserValves"): + __user__["valves"] = function_module.UserValves( + **Functions.get_user_valves_by_id_and_user_id( + filter_id, user.id + ) + ) + except Exception as e: + print(e) + + params = {**params, "__user__": __user__} + + if inspect.iscoroutinefunction(outlet): + data = await outlet(**params) + else: + data = outlet(**params) + + except Exception as e: + return Exception(f"Error: {e}") + + return data + + +async def chat_action(request: Request, action_id: str, form_data: dict, user: Any): + if "." in action_id: + action_id, sub_action_id = action_id.split(".") + else: + sub_action_id = None + + action = Functions.get_function_by_id(action_id) + if not action: + raise Exception(f"Action not found: {action_id}") + + await get_all_models(request) + models = request.app.state.MODELS + + data = form_data + model_id = data["model"] + + if model_id not in models: + raise Exception("Model not found") + model = models[model_id] + + __event_emitter__ = get_event_emitter( + { + "chat_id": data["chat_id"], + "message_id": data["id"], + "session_id": data["session_id"], + } + ) + __event_call__ = get_event_call( + { + "chat_id": data["chat_id"], + "message_id": data["id"], + "session_id": data["session_id"], + } + ) + + if action_id in request.app.state.FUNCTIONS: + function_module = request.app.state.FUNCTIONS[action_id] + else: + function_module, _, _ = load_function_module_by_id(action_id) + request.app.state.FUNCTIONS[action_id] = function_module + + if hasattr(function_module, "valves") and hasattr(function_module, "Valves"): + valves = Functions.get_function_valves_by_id(action_id) + function_module.valves = function_module.Valves(**(valves if valves else {})) + + if hasattr(function_module, "action"): + try: + action = function_module.action + + # Get the signature of the function + sig = inspect.signature(action) + params = {"body": data} + + # Extra parameters to be passed to the function + extra_params = { + "__model__": model, + "__id__": sub_action_id if sub_action_id is not None else action_id, + "__event_emitter__": __event_emitter__, + "__event_call__": __event_call__, + "__request__": request, + } + + # Add extra params in contained in function signature + for key, value in extra_params.items(): + if key in sig.parameters: + params[key] = value + + if "__user__" in sig.parameters: + __user__ = { + "id": user.id, + "email": user.email, + "name": user.name, + "role": user.role, + } + + try: + if hasattr(function_module, "UserValves"): + __user__["valves"] = function_module.UserValves( + **Functions.get_user_valves_by_id_and_user_id( + action_id, user.id + ) + ) + except Exception as e: + print(e) + + params = {**params, "__user__": __user__} + + if inspect.iscoroutinefunction(action): + data = await action(**params) + else: + data = action(**params) + + except Exception as e: + return Exception(f"Error: {e}") + + return data diff --git a/backend/open_webui/apps/images/utils/comfyui.py b/backend/open_webui/utils/images/comfyui.py similarity index 100% rename from backend/open_webui/apps/images/utils/comfyui.py rename to backend/open_webui/utils/images/comfyui.py diff --git a/backend/open_webui/utils/logo.png b/backend/open_webui/utils/logo.png deleted file mode 100644 index 519af1db6..000000000 Binary files a/backend/open_webui/utils/logo.png and /dev/null differ diff --git a/backend/open_webui/utils/middleware.py b/backend/open_webui/utils/middleware.py new file mode 100644 index 000000000..1d2bc2b99 --- /dev/null +++ b/backend/open_webui/utils/middleware.py @@ -0,0 +1,508 @@ +import time +import logging +import sys + +from aiocache import cached +from typing import Any, Optional +import random +import json +import inspect + +from fastapi import Request +from starlette.responses import Response, StreamingResponse + + +from open_webui.socket.main import ( + get_event_call, + get_event_emitter, +) +from open_webui.routers.tasks import generate_queries + + +from open_webui.models.users import UserModel +from open_webui.models.functions import Functions +from open_webui.models.models import Models + +from open_webui.retrieval.utils import get_sources_from_files + + +from open_webui.utils.chat import generate_chat_completion +from open_webui.utils.task import ( + get_task_model_id, + rag_template, + tools_function_calling_generation_template, +) +from open_webui.utils.misc import ( + add_or_update_system_message, + get_last_user_message, + prepend_to_first_user_message_content, +) +from open_webui.utils.tools import get_tools +from open_webui.utils.plugin import load_function_module_by_id + + +from open_webui.config import DEFAULT_TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE +from open_webui.env import SRC_LOG_LEVELS, GLOBAL_LOG_LEVEL, BYPASS_MODEL_ACCESS_CONTROL +from open_webui.constants import TASKS + + +logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL) +log = logging.getLogger(__name__) +log.setLevel(SRC_LOG_LEVELS["MAIN"]) + + +async def chat_completion_filter_functions_handler(request, body, model, extra_params): + skip_files = None + + def get_filter_function_ids(model): + def get_priority(function_id): + function = Functions.get_function_by_id(function_id) + if function is not None and hasattr(function, "valves"): + # TODO: Fix FunctionModel + return (function.valves if function.valves else {}).get("priority", 0) + return 0 + + filter_ids = [ + function.id for function in Functions.get_global_filter_functions() + ] + if "info" in model and "meta" in model["info"]: + filter_ids.extend(model["info"]["meta"].get("filterIds", [])) + filter_ids = list(set(filter_ids)) + + enabled_filter_ids = [ + function.id + for function in Functions.get_functions_by_type("filter", active_only=True) + ] + + filter_ids = [ + filter_id for filter_id in filter_ids if filter_id in enabled_filter_ids + ] + + filter_ids.sort(key=get_priority) + return filter_ids + + filter_ids = get_filter_function_ids(model) + for filter_id in filter_ids: + filter = Functions.get_function_by_id(filter_id) + if not filter: + continue + + if filter_id in request.app.state.FUNCTIONS: + function_module = request.app.state.FUNCTIONS[filter_id] + else: + function_module, _, _ = load_function_module_by_id(filter_id) + request.app.state.FUNCTIONS[filter_id] = function_module + + # Check if the function has a file_handler variable + if hasattr(function_module, "file_handler"): + skip_files = function_module.file_handler + + # Apply valves to the function + if hasattr(function_module, "valves") and hasattr(function_module, "Valves"): + valves = Functions.get_function_valves_by_id(filter_id) + function_module.valves = function_module.Valves( + **(valves if valves else {}) + ) + + if hasattr(function_module, "inlet"): + try: + inlet = function_module.inlet + + # Create a dictionary of parameters to be passed to the function + params = {"body": body} | { + k: v + for k, v in { + **extra_params, + "__model__": model, + "__id__": filter_id, + }.items() + if k in inspect.signature(inlet).parameters + } + + if "__user__" in params and hasattr(function_module, "UserValves"): + try: + params["__user__"]["valves"] = function_module.UserValves( + **Functions.get_user_valves_by_id_and_user_id( + filter_id, params["__user__"]["id"] + ) + ) + except Exception as e: + print(e) + + if inspect.iscoroutinefunction(inlet): + body = await inlet(**params) + else: + body = inlet(**params) + + except Exception as e: + print(f"Error: {e}") + raise e + + if skip_files and "files" in body.get("metadata", {}): + del body["metadata"]["files"] + + return body, {} + + +async def chat_completion_tools_handler( + request: Request, body: dict, user: UserModel, models, extra_params: dict +) -> tuple[dict, dict]: + async def get_content_from_response(response) -> Optional[str]: + content = None + if hasattr(response, "body_iterator"): + async for chunk in response.body_iterator: + data = json.loads(chunk.decode("utf-8")) + content = data["choices"][0]["message"]["content"] + + # Cleanup any remaining background tasks if necessary + if response.background is not None: + await response.background() + else: + content = response["choices"][0]["message"]["content"] + return content + + def get_tools_function_calling_payload(messages, task_model_id, content): + user_message = get_last_user_message(messages) + history = "\n".join( + f"{message['role'].upper()}: \"\"\"{message['content']}\"\"\"" + for message in messages[::-1][:4] + ) + + prompt = f"History:\n{history}\nQuery: {user_message}" + + return { + "model": task_model_id, + "messages": [ + {"role": "system", "content": content}, + {"role": "user", "content": f"Query: {prompt}"}, + ], + "stream": False, + "metadata": {"task": str(TASKS.FUNCTION_CALLING)}, + } + + # If tool_ids field is present, call the functions + metadata = body.get("metadata", {}) + + tool_ids = metadata.get("tool_ids", None) + log.debug(f"{tool_ids=}") + if not tool_ids: + return body, {} + + skip_files = False + sources = [] + + task_model_id = get_task_model_id( + body["model"], + request.app.state.config.TASK_MODEL, + request.app.state.config.TASK_MODEL_EXTERNAL, + models, + ) + tools = get_tools( + request, + tool_ids, + user, + { + **extra_params, + "__model__": models[task_model_id], + "__messages__": body["messages"], + "__files__": metadata.get("files", []), + }, + ) + log.info(f"{tools=}") + + specs = [tool["spec"] for tool in tools.values()] + tools_specs = json.dumps(specs) + + if request.app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE != "": + template = request.app.state.config.TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE + else: + template = DEFAULT_TOOLS_FUNCTION_CALLING_PROMPT_TEMPLATE + + tools_function_calling_prompt = tools_function_calling_generation_template( + template, tools_specs + ) + log.info(f"{tools_function_calling_prompt=}") + payload = get_tools_function_calling_payload( + body["messages"], task_model_id, tools_function_calling_prompt + ) + + try: + response = await generate_chat_completion(request, form_data=payload, user=user) + log.debug(f"{response=}") + content = await get_content_from_response(response) + log.debug(f"{content=}") + + if not content: + return body, {} + + try: + content = content[content.find("{") : content.rfind("}") + 1] + if not content: + raise Exception("No JSON object found in the response") + + result = json.loads(content) + + tool_function_name = result.get("name", None) + if tool_function_name not in tools: + return body, {} + + tool_function_params = result.get("parameters", {}) + + try: + required_params = ( + tools[tool_function_name] + .get("spec", {}) + .get("parameters", {}) + .get("required", []) + ) + tool_function = tools[tool_function_name]["callable"] + tool_function_params = { + k: v + for k, v in tool_function_params.items() + if k in required_params + } + tool_output = await tool_function(**tool_function_params) + + except Exception as e: + tool_output = str(e) + + if isinstance(tool_output, str): + if tools[tool_function_name]["citation"]: + sources.append( + { + "source": { + "name": f"TOOL:{tools[tool_function_name]['toolkit_id']}/{tool_function_name}" + }, + "document": [tool_output], + "metadata": [ + { + "source": f"TOOL:{tools[tool_function_name]['toolkit_id']}/{tool_function_name}" + } + ], + } + ) + else: + sources.append( + { + "source": {}, + "document": [tool_output], + "metadata": [ + { + "source": f"TOOL:{tools[tool_function_name]['toolkit_id']}/{tool_function_name}" + } + ], + } + ) + + if tools[tool_function_name]["file_handler"]: + skip_files = True + + except Exception as e: + log.exception(f"Error: {e}") + content = None + except Exception as e: + log.exception(f"Error: {e}") + content = None + + log.debug(f"tool_contexts: {sources}") + + if skip_files and "files" in body.get("metadata", {}): + del body["metadata"]["files"] + + return body, {"sources": sources} + + +async def chat_completion_files_handler( + request: Request, body: dict, user: UserModel +) -> tuple[dict, dict[str, list]]: + sources = [] + + if files := body.get("metadata", {}).get("files", None): + try: + queries_response = await generate_queries( + { + "model": body["model"], + "messages": body["messages"], + "type": "retrieval", + }, + user, + ) + queries_response = queries_response["choices"][0]["message"]["content"] + + try: + bracket_start = queries_response.find("{") + bracket_end = queries_response.rfind("}") + 1 + + if bracket_start == -1 or bracket_end == -1: + raise Exception("No JSON object found in the response") + + queries_response = queries_response[bracket_start:bracket_end] + queries_response = json.loads(queries_response) + except Exception as e: + queries_response = {"queries": [queries_response]} + + queries = queries_response.get("queries", []) + except Exception as e: + queries = [] + + if len(queries) == 0: + queries = [get_last_user_message(body["messages"])] + + sources = get_sources_from_files( + files=files, + queries=queries, + embedding_function=request.app.state.EMBEDDING_FUNCTION, + k=request.app.state.config.TOP_K, + reranking_function=request.app.state.rf, + r=request.app.state.config.RELEVANCE_THRESHOLD, + hybrid_search=request.app.state.config.ENABLE_RAG_HYBRID_SEARCH, + ) + + log.debug(f"rag_contexts:sources: {sources}") + return body, {"sources": sources} + + +async def process_chat_payload(request, form_data, user, model): + metadata = { + "chat_id": form_data.pop("chat_id", None), + "message_id": form_data.pop("id", None), + "session_id": form_data.pop("session_id", None), + "tool_ids": form_data.get("tool_ids", None), + "files": form_data.get("files", None), + } + form_data["metadata"] = metadata + + extra_params = { + "__event_emitter__": get_event_emitter(metadata), + "__event_call__": get_event_call(metadata), + "__user__": { + "id": user.id, + "email": user.email, + "name": user.name, + "role": user.role, + }, + "__metadata__": metadata, + "__request__": request, + } + + # Initialize events to store additional event to be sent to the client + # Initialize contexts and citation + models = request.app.state.MODELS + events = [] + sources = [] + + try: + form_data, flags = await chat_completion_filter_functions_handler( + request, form_data, model, extra_params + ) + except Exception as e: + return Exception(f"Error: {e}") + + tool_ids = form_data.pop("tool_ids", None) + files = form_data.pop("files", None) + + metadata = { + **metadata, + "tool_ids": tool_ids, + "files": files, + } + form_data["metadata"] = metadata + + try: + form_data, flags = await chat_completion_tools_handler( + request, form_data, user, models, extra_params + ) + sources.extend(flags.get("sources", [])) + except Exception as e: + log.exception(e) + + try: + form_data, flags = await chat_completion_files_handler(request, form_data, user) + sources.extend(flags.get("sources", [])) + except Exception as e: + log.exception(e) + + # If context is not empty, insert it into the messages + if len(sources) > 0: + context_string = "" + for source_idx, source in enumerate(sources): + source_id = source.get("source", {}).get("name", "") + + if "document" in source: + for doc_idx, doc_context in enumerate(source["document"]): + metadata = source.get("metadata") + doc_source_id = None + + if metadata: + doc_source_id = metadata[doc_idx].get("source", source_id) + + if source_id: + context_string += f"{doc_source_id if doc_source_id is not None else source_id}{doc_context}\n" + else: + # If there is no source_id, then do not include the source_id tag + context_string += f"{doc_context}\n" + + context_string = context_string.strip() + prompt = get_last_user_message(form_data["messages"]) + + if prompt is None: + raise Exception("No user message found") + if ( + request.app.state.config.RELEVANCE_THRESHOLD == 0 + and context_string.strip() == "" + ): + log.debug( + f"With a 0 relevancy threshold for RAG, the context cannot be empty" + ) + + # Workaround for Ollama 2.0+ system prompt issue + # TODO: replace with add_or_update_system_message + if model["owned_by"] == "ollama": + form_data["messages"] = prepend_to_first_user_message_content( + rag_template( + request.app.state.config.RAG_TEMPLATE, context_string, prompt + ), + form_data["messages"], + ) + else: + form_data["messages"] = add_or_update_system_message( + rag_template( + request.app.state.config.RAG_TEMPLATE, context_string, prompt + ), + form_data["messages"], + ) + + # If there are citations, add them to the data_items + sources = [source for source in sources if source.get("source", {}).get("name", "")] + + if len(sources) > 0: + events.append({"sources": sources}) + + return form_data, events + + +async def process_chat_response(response, events): + if not isinstance(response, StreamingResponse): + return response + + content_type = response.headers["Content-Type"] + is_openai = "text/event-stream" in content_type + is_ollama = "application/x-ndjson" in content_type + + if not is_openai and not is_ollama: + return response + + async def stream_wrapper(original_generator, events): + def wrap_item(item): + return f"data: {item}\n\n" if is_openai else f"{item}\n" + + for event in events: + yield wrap_item(json.dumps(event)) + + async for data in original_generator: + yield data + + return StreamingResponse( + stream_wrapper(response.body_iterator, events), + headers=dict(response.headers), + ) diff --git a/backend/open_webui/utils/misc.py b/backend/open_webui/utils/misc.py index a5af492ba..aba696f60 100644 --- a/backend/open_webui/utils/misc.py +++ b/backend/open_webui/utils/misc.py @@ -106,7 +106,7 @@ def openai_chat_message_template(model: str): def openai_chat_chunk_message_template( - model: str, message: Optional[str] = None + model: str, message: Optional[str] = None, usage: Optional[dict] = None ) -> dict: template = openai_chat_message_template(model) template["object"] = "chat.completion.chunk" @@ -114,17 +114,23 @@ def openai_chat_chunk_message_template( template["choices"][0]["delta"] = {"content": message} else: template["choices"][0]["finish_reason"] = "stop" + + if usage: + template["usage"] = usage return template def openai_chat_completion_message_template( - model: str, message: Optional[str] = None + model: str, message: Optional[str] = None, usage: Optional[dict] = None ) -> dict: template = openai_chat_message_template(model) template["object"] = "chat.completion" if message is not None: template["choices"][0]["message"] = {"content": message, "role": "assistant"} template["choices"][0]["finish_reason"] = "stop" + + if usage: + template["usage"] = usage return template diff --git a/backend/open_webui/utils/models.py b/backend/open_webui/utils/models.py new file mode 100644 index 000000000..b135f8997 --- /dev/null +++ b/backend/open_webui/utils/models.py @@ -0,0 +1,246 @@ +import time +import logging +import sys + +from aiocache import cached +from fastapi import Request + +from open_webui.routers import openai, ollama +from open_webui.functions import get_function_models + + +from open_webui.models.functions import Functions +from open_webui.models.models import Models + + +from open_webui.utils.plugin import load_function_module_by_id +from open_webui.utils.access_control import has_access + + +from open_webui.config import ( + DEFAULT_ARENA_MODEL, +) + +from open_webui.env import SRC_LOG_LEVELS, GLOBAL_LOG_LEVEL + + +logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL) +log = logging.getLogger(__name__) +log.setLevel(SRC_LOG_LEVELS["MAIN"]) + + +async def get_all_base_models(request: Request): + function_models = [] + openai_models = [] + ollama_models = [] + + if request.app.state.config.ENABLE_OPENAI_API: + openai_models = await openai.get_all_models(request) + openai_models = openai_models["data"] + + if request.app.state.config.ENABLE_OLLAMA_API: + ollama_models = await ollama.get_all_models(request) + ollama_models = [ + { + "id": model["model"], + "name": model["name"], + "object": "model", + "created": int(time.time()), + "owned_by": "ollama", + "ollama": model, + } + for model in ollama_models["models"] + ] + + function_models = await get_function_models() + models = function_models + openai_models + ollama_models + + return models + + +@cached(ttl=3) +async def get_all_models(request): + models = await get_all_base_models(request) + + # If there are no models, return an empty list + if len(models) == 0: + return [] + + # Add arena models + if request.app.state.config.ENABLE_EVALUATION_ARENA_MODELS: + arena_models = [] + if len(request.app.state.config.EVALUATION_ARENA_MODELS) > 0: + arena_models = [ + { + "id": model["id"], + "name": model["name"], + "info": { + "meta": model["meta"], + }, + "object": "model", + "created": int(time.time()), + "owned_by": "arena", + "arena": True, + } + for model in request.app.state.config.EVALUATION_ARENA_MODELS + ] + else: + # Add default arena model + arena_models = [ + { + "id": DEFAULT_ARENA_MODEL["id"], + "name": DEFAULT_ARENA_MODEL["name"], + "info": { + "meta": DEFAULT_ARENA_MODEL["meta"], + }, + "object": "model", + "created": int(time.time()), + "owned_by": "arena", + "arena": True, + } + ] + models = models + arena_models + + global_action_ids = [ + function.id for function in Functions.get_global_action_functions() + ] + enabled_action_ids = [ + function.id + for function in Functions.get_functions_by_type("action", active_only=True) + ] + + custom_models = Models.get_all_models() + for custom_model in custom_models: + if custom_model.base_model_id is None: + for model in models: + if ( + custom_model.id == model["id"] + or custom_model.id == model["id"].split(":")[0] + ): + if custom_model.is_active: + model["name"] = custom_model.name + model["info"] = custom_model.model_dump() + + action_ids = [] + if "info" in model and "meta" in model["info"]: + action_ids.extend( + model["info"]["meta"].get("actionIds", []) + ) + + model["action_ids"] = action_ids + else: + models.remove(model) + + elif custom_model.is_active and ( + custom_model.id not in [model["id"] for model in models] + ): + owned_by = "openai" + pipe = None + action_ids = [] + + for model in models: + if ( + custom_model.base_model_id == model["id"] + or custom_model.base_model_id == model["id"].split(":")[0] + ): + owned_by = model["owned_by"] + if "pipe" in model: + pipe = model["pipe"] + break + + if custom_model.meta: + meta = custom_model.meta.model_dump() + if "actionIds" in meta: + action_ids.extend(meta["actionIds"]) + + models.append( + { + "id": f"{custom_model.id}", + "name": custom_model.name, + "object": "model", + "created": custom_model.created_at, + "owned_by": owned_by, + "info": custom_model.model_dump(), + "preset": True, + **({"pipe": pipe} if pipe is not None else {}), + "action_ids": action_ids, + } + ) + + # Process action_ids to get the actions + def get_action_items_from_module(function, module): + actions = [] + if hasattr(module, "actions"): + actions = module.actions + return [ + { + "id": f"{function.id}.{action['id']}", + "name": action.get("name", f"{function.name} ({action['id']})"), + "description": function.meta.description, + "icon_url": action.get( + "icon_url", function.meta.manifest.get("icon_url", None) + ), + } + for action in actions + ] + else: + return [ + { + "id": function.id, + "name": function.name, + "description": function.meta.description, + "icon_url": function.meta.manifest.get("icon_url", None), + } + ] + + def get_function_module_by_id(function_id): + if function_id in request.app.state.FUNCTIONS: + function_module = request.app.state.FUNCTIONS[function_id] + else: + function_module, _, _ = load_function_module_by_id(function_id) + request.app.state.FUNCTIONS[function_id] = function_module + + for model in models: + action_ids = [ + action_id + for action_id in list(set(model.pop("action_ids", []) + global_action_ids)) + if action_id in enabled_action_ids + ] + + model["actions"] = [] + for action_id in action_ids: + action_function = Functions.get_function_by_id(action_id) + if action_function is None: + raise Exception(f"Action not found: {action_id}") + + function_module = get_function_module_by_id(action_id) + model["actions"].extend( + get_action_items_from_module(action_function, function_module) + ) + log.debug(f"get_all_models() returned {len(models)} models") + + request.app.state.MODELS = {model["id"]: model for model in models} + return models + + +def check_model_access(user, model): + if model.get("arena"): + if not has_access( + user.id, + type="read", + access_control=model.get("info", {}) + .get("meta", {}) + .get("access_control", {}), + ): + raise Exception("Model not found") + else: + model_info = Models.get_model_by_id(model.get("id")) + if not model_info: + raise Exception("Model not found") + elif not ( + user.id == model_info.user_id + or has_access( + user.id, type="read", access_control=model_info.access_control + ) + ): + raise Exception("Model not found") diff --git a/backend/open_webui/utils/oauth.py b/backend/open_webui/utils/oauth.py index 37dc5b788..f0ab7a345 100644 --- a/backend/open_webui/utils/oauth.py +++ b/backend/open_webui/utils/oauth.py @@ -12,8 +12,8 @@ from fastapi import ( ) from starlette.responses import RedirectResponse -from open_webui.apps.webui.models.auths import Auths -from open_webui.apps.webui.models.users import Users +from open_webui.models.auths import Auths +from open_webui.models.users import Users from open_webui.config import ( DEFAULT_USER_ROLE, ENABLE_OAUTH_SIGNUP, @@ -158,8 +158,13 @@ class OAuthManager: if not email: log.warning(f"OAuth callback failed, email is missing: {user_data}") raise HTTPException(400, detail=ERROR_MESSAGES.INVALID_CRED) - if "*" not in auth_manager_config.OAUTH_ALLOWED_DOMAINS and email.split("@")[-1] not in auth_manager_config.OAUTH_ALLOWED_DOMAINS: - log.warning(f"OAuth callback failed, e-mail domain is not in the list of allowed domains: {user_data}") + if ( + "*" not in auth_manager_config.OAUTH_ALLOWED_DOMAINS + and email.split("@")[-1] not in auth_manager_config.OAUTH_ALLOWED_DOMAINS + ): + log.warning( + f"OAuth callback failed, e-mail domain is not in the list of allowed domains: {user_data}" + ) raise HTTPException(400, detail=ERROR_MESSAGES.INVALID_CRED) # Check if the user exists diff --git a/backend/open_webui/utils/pdf_generator.py b/backend/open_webui/utils/pdf_generator.py index 6b7d506e6..bbaf42dbb 100644 --- a/backend/open_webui/utils/pdf_generator.py +++ b/backend/open_webui/utils/pdf_generator.py @@ -9,7 +9,7 @@ import site from fpdf import FPDF from open_webui.env import STATIC_DIR, FONTS_DIR -from open_webui.apps.webui.models.chats import ChatTitleMessagesForm +from open_webui.models.chats import ChatTitleMessagesForm class PDFGenerator: diff --git a/backend/open_webui/apps/webui/utils.py b/backend/open_webui/utils/plugin.py similarity index 98% rename from backend/open_webui/apps/webui/utils.py rename to backend/open_webui/utils/plugin.py index 054158b3e..17b86cea1 100644 --- a/backend/open_webui/apps/webui/utils.py +++ b/backend/open_webui/utils/plugin.py @@ -8,8 +8,8 @@ import tempfile import logging from open_webui.env import SRC_LOG_LEVELS -from open_webui.apps.webui.models.functions import Functions -from open_webui.apps.webui.models.tools import Tools +from open_webui.models.functions import Functions +from open_webui.models.tools import Tools log = logging.getLogger(__name__) log.setLevel(SRC_LOG_LEVELS["MAIN"]) diff --git a/backend/open_webui/utils/response.py b/backend/open_webui/utils/response.py index b8501e92c..891016e43 100644 --- a/backend/open_webui/utils/response.py +++ b/backend/open_webui/utils/response.py @@ -21,8 +21,63 @@ async def convert_streaming_response_ollama_to_openai(ollama_streaming_response) message_content = data.get("message", {}).get("content", "") done = data.get("done", False) + usage = None + if done: + usage = { + "response_token/s": ( + round( + ( + ( + data.get("eval_count", 0) + / ((data.get("eval_duration", 0) / 1_000_000_000)) + ) + * 100 + ), + 2, + ) + if data.get("eval_duration", 0) > 0 + else "N/A" + ), + "prompt_token/s": ( + round( + ( + ( + data.get("prompt_eval_count", 0) + / ( + ( + data.get("prompt_eval_duration", 0) + / 1_000_000_000 + ) + ) + ) + * 100 + ), + 2, + ) + if data.get("prompt_eval_duration", 0) > 0 + else "N/A" + ), + "total_duration": round( + ((data.get("total_duration", 0) / 1_000_000) * 100), 2 + ), + "load_duration": round( + ((data.get("load_duration", 0) / 1_000_000) * 100), 2 + ), + "prompt_eval_count": data.get("prompt_eval_count", 0), + "prompt_eval_duration": round( + ((data.get("prompt_eval_duration", 0) / 1_000_000) * 100), 2 + ), + "eval_count": data.get("eval_count", 0), + "eval_duration": round( + ((data.get("eval_duration", 0) / 1_000_000) * 100), 2 + ), + "approximate_total": ( + lambda s: f"{s // 3600}h{(s % 3600) // 60}m{s % 60}s" + )((data.get("total_duration", 0) or 0) // 1_000_000_000), + } + data = openai_chat_chunk_message_template( - model, message_content if not done else None + model, message_content if not done else None, usage ) line = f"data: {json.dumps(data)}\n\n" diff --git a/backend/open_webui/utils/task.py b/backend/open_webui/utils/task.py index 604161a31..ebb7483ba 100644 --- a/backend/open_webui/utils/task.py +++ b/backend/open_webui/utils/task.py @@ -16,6 +16,22 @@ log = logging.getLogger(__name__) log.setLevel(SRC_LOG_LEVELS["RAG"]) +def get_task_model_id( + default_model_id: str, task_model: str, task_model_external: str, models +) -> str: + # Set the task model + task_model_id = default_model_id + # Check if the user has a custom task model and use that model + if models[task_model_id]["owned_by"] == "ollama": + if task_model and task_model in models: + task_model_id = task_model + else: + if task_model_external and task_model_external in models: + task_model_id = task_model_external + + return task_model_id + + def prompt_template( template: str, user_name: Optional[str] = None, user_location: Optional[str] = None ) -> str: diff --git a/backend/open_webui/utils/tools.py b/backend/open_webui/utils/tools.py index 60a9f942f..b6e13011d 100644 --- a/backend/open_webui/utils/tools.py +++ b/backend/open_webui/utils/tools.py @@ -4,11 +4,15 @@ import re from typing import Any, Awaitable, Callable, get_type_hints from functools import update_wrapper, partial -from langchain_core.utils.function_calling import convert_to_openai_function -from open_webui.apps.webui.models.tools import Tools -from open_webui.apps.webui.models.users import UserModel -from open_webui.apps.webui.utils import load_tools_module_by_id + +from fastapi import Request from pydantic import BaseModel, Field, create_model +from langchain_core.utils.function_calling import convert_to_openai_function + + +from open_webui.models.tools import Tools +from open_webui.models.users import UserModel +from open_webui.utils.plugin import load_tools_module_by_id log = logging.getLogger(__name__) @@ -32,7 +36,7 @@ def apply_extra_params_to_tool_function( # Mutation on extra_params def get_tools( - webui_app, tool_ids: list[str], user: UserModel, extra_params: dict + request: Request, tool_ids: list[str], user: UserModel, extra_params: dict ) -> dict[str, dict]: tools_dict = {} @@ -41,10 +45,10 @@ def get_tools( if tools is None: continue - module = webui_app.state.TOOLS.get(tool_id, None) + module = request.app.state.TOOLS.get(tool_id, None) if module is None: module, _ = load_tools_module_by_id(tool_id) - webui_app.state.TOOLS[tool_id] = module + request.app.state.TOOLS[tool_id] = module extra_params["__id__"] = tool_id if hasattr(module, "valves") and hasattr(module, "Valves"): diff --git a/package-lock.json b/package-lock.json index 020cd0f53..16542ed99 100644 --- a/package-lock.json +++ b/package-lock.json @@ -27,6 +27,7 @@ "async": "^3.2.5", "bits-ui": "^0.19.7", "codemirror": "^6.0.1", + "codemirror-lang-hcl": "^0.0.0-beta.2", "crc-32": "^1.2.2", "dayjs": "^1.11.10", "dompurify": "^3.1.6", @@ -4267,6 +4268,17 @@ "@codemirror/view": "^6.0.0" } }, + "node_modules/codemirror-lang-hcl": { + "version": "0.0.0-beta.2", + "resolved": "https://registry.npmjs.org/codemirror-lang-hcl/-/codemirror-lang-hcl-0.0.0-beta.2.tgz", + "integrity": "sha512-R3ew7Z2EYTdHTMXsWKBW9zxnLoLPYO+CrAa3dPZjXLrIR96Q3GR4cwJKF7zkSsujsnWgwRQZonyWpXYXfhQYuQ==", + "license": "MIT", + "dependencies": { + "@codemirror/language": "^6.0.0", + "@lezer/highlight": "^1.0.0", + "@lezer/lr": "^1.0.0" + } + }, "node_modules/coincident": { "version": "1.2.3", "resolved": "https://registry.npmjs.org/coincident/-/coincident-1.2.3.tgz", diff --git a/package.json b/package.json index c131e1f91..3b5911791 100644 --- a/package.json +++ b/package.json @@ -50,6 +50,7 @@ "type": "module", "dependencies": { "@codemirror/lang-javascript": "^6.2.2", + "codemirror-lang-hcl": "^0.0.0-beta.2", "@codemirror/lang-python": "^6.1.6", "@codemirror/language-data": "^6.5.1", "@codemirror/theme-one-dark": "^6.1.2", diff --git a/pyproject.toml b/pyproject.toml index 0554baa9e..de14a9fa1 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -105,13 +105,14 @@ dependencies = [ "ldap3==2.9.1" ] readme = "README.md" -requires-python = ">= 3.11, < 3.12.0a1" +requires-python = ">= 3.11, < 3.13.0a1" dynamic = ["version"] classifiers = [ "Development Status :: 4 - Beta", "License :: OSI Approved :: MIT License", "Programming Language :: Python :: 3", "Programming Language :: Python :: 3.11", + "Programming Language :: Python :: 3.12", "Topic :: Communications :: Chat", "Topic :: Multimedia", ] diff --git a/src/lib/apis/index.ts b/src/lib/apis/index.ts index e76aa3c99..d06fbf3d7 100644 --- a/src/lib/apis/index.ts +++ b/src/lib/apis/index.ts @@ -110,7 +110,7 @@ export const chatAction = async (token: string, action_id: string, body: ChatAct export const getTaskConfig = async (token: string = '') => { let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/task/config`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/tasks/config`, { method: 'GET', headers: { Accept: 'application/json', @@ -138,7 +138,7 @@ export const getTaskConfig = async (token: string = '') => { export const updateTaskConfig = async (token: string, config: object) => { let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/task/config/update`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/tasks/config/update`, { method: 'POST', headers: { Accept: 'application/json', @@ -176,7 +176,7 @@ export const generateTitle = async ( ) => { let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/task/title/completions`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/tasks/title/completions`, { method: 'POST', headers: { Accept: 'application/json', @@ -216,7 +216,7 @@ export const generateTags = async ( ) => { let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/task/tags/completions`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/tasks/tags/completions`, { method: 'POST', headers: { Accept: 'application/json', @@ -288,7 +288,7 @@ export const generateEmoji = async ( ) => { let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/task/emoji/completions`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/tasks/emoji/completions`, { method: 'POST', headers: { Accept: 'application/json', @@ -337,7 +337,7 @@ export const generateQueries = async ( ) => { let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/task/queries/completions`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/tasks/queries/completions`, { method: 'POST', headers: { Accept: 'application/json', @@ -407,7 +407,7 @@ export const generateAutoCompletion = async ( const controller = new AbortController(); let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/task/auto/completions`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/tasks/auto/completions`, { signal: controller.signal, method: 'POST', headers: { @@ -477,7 +477,7 @@ export const generateMoACompletion = async ( const controller = new AbortController(); let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/task/moa/completions`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/tasks/moa/completions`, { signal: controller.signal, method: 'POST', headers: { @@ -507,7 +507,7 @@ export const generateMoACompletion = async ( export const getPipelinesList = async (token: string = '') => { let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/pipelines/list`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/pipelines/list`, { method: 'GET', headers: { Accept: 'application/json', @@ -541,7 +541,7 @@ export const uploadPipeline = async (token: string, file: File, urlIdx: string) formData.append('file', file); formData.append('urlIdx', urlIdx); - const res = await fetch(`${WEBUI_BASE_URL}/api/pipelines/upload`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/pipelines/upload`, { method: 'POST', headers: { ...(token && { authorization: `Bearer ${token}` }) @@ -573,7 +573,7 @@ export const uploadPipeline = async (token: string, file: File, urlIdx: string) export const downloadPipeline = async (token: string, url: string, urlIdx: string) => { let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/pipelines/add`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/pipelines/add`, { method: 'POST', headers: { Accept: 'application/json', @@ -609,7 +609,7 @@ export const downloadPipeline = async (token: string, url: string, urlIdx: strin export const deletePipeline = async (token: string, id: string, urlIdx: string) => { let error = null; - const res = await fetch(`${WEBUI_BASE_URL}/api/pipelines/delete`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/pipelines/delete`, { method: 'DELETE', headers: { Accept: 'application/json', @@ -650,7 +650,7 @@ export const getPipelines = async (token: string, urlIdx?: string) => { searchParams.append('urlIdx', urlIdx); } - const res = await fetch(`${WEBUI_BASE_URL}/api/pipelines?${searchParams.toString()}`, { + const res = await fetch(`${WEBUI_BASE_URL}/api/v1/pipelines?${searchParams.toString()}`, { method: 'GET', headers: { Accept: 'application/json', @@ -685,7 +685,7 @@ export const getPipelineValves = async (token: string, pipeline_id: string, urlI } const res = await fetch( - `${WEBUI_BASE_URL}/api/pipelines/${pipeline_id}/valves?${searchParams.toString()}`, + `${WEBUI_BASE_URL}/api/v1/pipelines/${pipeline_id}/valves?${searchParams.toString()}`, { method: 'GET', headers: { @@ -721,7 +721,7 @@ export const getPipelineValvesSpec = async (token: string, pipeline_id: string, } const res = await fetch( - `${WEBUI_BASE_URL}/api/pipelines/${pipeline_id}/valves/spec?${searchParams.toString()}`, + `${WEBUI_BASE_URL}/api/v1/pipelines/${pipeline_id}/valves/spec?${searchParams.toString()}`, { method: 'GET', headers: { @@ -762,7 +762,7 @@ export const updatePipelineValves = async ( } const res = await fetch( - `${WEBUI_BASE_URL}/api/pipelines/${pipeline_id}/valves/update?${searchParams.toString()}`, + `${WEBUI_BASE_URL}/api/v1/pipelines/${pipeline_id}/valves/update?${searchParams.toString()}`, { method: 'POST', headers: { diff --git a/src/lib/apis/streaming/index.ts b/src/lib/apis/streaming/index.ts index 54804385d..5617ce36c 100644 --- a/src/lib/apis/streaming/index.ts +++ b/src/lib/apis/streaming/index.ts @@ -77,10 +77,14 @@ async function* openAIStreamToIterator( continue; } + if (parsedData.usage) { + yield { done: false, value: '', usage: parsedData.usage }; + continue; + } + yield { done: false, value: parsedData.choices?.[0]?.delta?.content ?? '', - usage: parsedData.usage }; } catch (e) { console.error('Error extracting delta from SSE event:', e); @@ -98,10 +102,26 @@ async function* streamLargeDeltasAsRandomChunks( yield textStreamUpdate; return; } + + if (textStreamUpdate.error) { + yield textStreamUpdate; + continue; + } if (textStreamUpdate.sources) { yield textStreamUpdate; continue; } + if (textStreamUpdate.selectedModelId) { + yield textStreamUpdate; + continue; + } + if (textStreamUpdate.usage) { + yield textStreamUpdate; + continue; + } + + + let content = textStreamUpdate.value; if (content.length < 5) { yield { done: false, value: content }; diff --git a/src/lib/components/chat/Chat.svelte b/src/lib/components/chat/Chat.svelte index e6a653420..a55cbc87b 100644 --- a/src/lib/components/chat/Chat.svelte +++ b/src/lib/components/chat/Chat.svelte @@ -455,41 +455,43 @@ ////////////////////////// const initNewChat = async () => { - if (sessionStorage.selectedModels) { - selectedModels = JSON.parse(sessionStorage.selectedModels); - sessionStorage.removeItem('selectedModels'); - } else { - if ($page.url.searchParams.get('models')) { - selectedModels = $page.url.searchParams.get('models')?.split(','); - } else if ($page.url.searchParams.get('model')) { - const urlModels = $page.url.searchParams.get('model')?.split(','); + if ($page.url.searchParams.get('models')) { + selectedModels = $page.url.searchParams.get('models')?.split(','); + } else if ($page.url.searchParams.get('model')) { + const urlModels = $page.url.searchParams.get('model')?.split(','); - if (urlModels.length === 1) { - const m = $models.find((m) => m.id === urlModels[0]); - if (!m) { - const modelSelectorButton = document.getElementById('model-selector-0-button'); - if (modelSelectorButton) { - modelSelectorButton.click(); - await tick(); + if (urlModels.length === 1) { + const m = $models.find((m) => m.id === urlModels[0]); + if (!m) { + const modelSelectorButton = document.getElementById('model-selector-0-button'); + if (modelSelectorButton) { + modelSelectorButton.click(); + await tick(); - const modelSelectorInput = document.getElementById('model-search-input'); - if (modelSelectorInput) { - modelSelectorInput.focus(); - modelSelectorInput.value = urlModels[0]; - modelSelectorInput.dispatchEvent(new Event('input')); - } + const modelSelectorInput = document.getElementById('model-search-input'); + if (modelSelectorInput) { + modelSelectorInput.focus(); + modelSelectorInput.value = urlModels[0]; + modelSelectorInput.dispatchEvent(new Event('input')); } - } else { - selectedModels = urlModels; } } else { selectedModels = urlModels; } - } else if ($settings?.models) { - selectedModels = $settings?.models; - } else if ($config?.default_models) { - console.log($config?.default_models.split(',') ?? ''); - selectedModels = $config?.default_models.split(','); + } else { + selectedModels = urlModels; + } + } else { + if (sessionStorage.selectedModels) { + selectedModels = JSON.parse(sessionStorage.selectedModels); + sessionStorage.removeItem('selectedModels'); + } else { + if ($settings?.models) { + selectedModels = $settings?.models; + } else if ($config?.default_models) { + console.log($config?.default_models.split(',') ?? ''); + selectedModels = $config?.default_models.split(','); + } } } @@ -1056,11 +1058,14 @@ } let _response = null; - if (model?.owned_by === 'ollama') { - _response = await sendPromptOllama(model, prompt, responseMessageId, _chatId); - } else if (model) { - _response = await sendPromptOpenAI(model, prompt, responseMessageId, _chatId); - } + + // if (model?.owned_by === 'ollama') { + // _response = await sendPromptOllama(model, prompt, responseMessageId, _chatId); + // } else if (model) { + // } + + _response = await sendPromptOpenAI(model, prompt, responseMessageId, _chatId); + _responses.push(_response); if (chatEventEmitter) clearInterval(chatEventEmitter); @@ -1207,24 +1212,14 @@ $settings?.params?.stream_response ?? params?.stream_response ?? true; + const [res, controller] = await generateChatCompletion(localStorage.token, { stream: stream, model: model.id, messages: messagesBody, - options: { - ...{ ...($settings?.params ?? {}), ...params }, - stop: - (params?.stop ?? $settings?.params?.stop ?? undefined) - ? (params?.stop.split(',').map((token) => token.trim()) ?? $settings.params.stop).map( - (str) => decodeURIComponent(JSON.parse('"' + str.replace(/\"/g, '\\"') + '"')) - ) - : undefined, - num_predict: params?.max_tokens ?? $settings?.params?.max_tokens ?? undefined, - repeat_penalty: - params?.frequency_penalty ?? $settings?.params?.frequency_penalty ?? undefined - }, format: $settings.requestFormat ?? undefined, keep_alive: $settings.keepAlive ?? undefined, + tool_ids: selectedToolIds.length > 0 ? selectedToolIds : undefined, files: files.length > 0 ? files : undefined, session_id: $socket?.id, @@ -1542,13 +1537,6 @@ { stream: stream, model: model.id, - ...(stream && (model.info?.meta?.capabilities?.usage ?? false) - ? { - stream_options: { - include_usage: true - } - } - : {}), messages: [ params?.system || $settings.system || (responseMessage?.userContext ?? null) ? { @@ -1593,23 +1581,36 @@ content: message?.merged?.content ?? message.content }) })), - seed: params?.seed ?? $settings?.params?.seed ?? undefined, - stop: - (params?.stop ?? $settings?.params?.stop ?? undefined) - ? (params?.stop.split(',').map((token) => token.trim()) ?? $settings.params.stop).map( - (str) => decodeURIComponent(JSON.parse('"' + str.replace(/\"/g, '\\"') + '"')) - ) - : undefined, - temperature: params?.temperature ?? $settings?.params?.temperature ?? undefined, - top_p: params?.top_p ?? $settings?.params?.top_p ?? undefined, - frequency_penalty: - params?.frequency_penalty ?? $settings?.params?.frequency_penalty ?? undefined, - max_tokens: params?.max_tokens ?? $settings?.params?.max_tokens ?? undefined, + + // params: { + // ...$settings?.params, + // ...params, + + // format: $settings.requestFormat ?? undefined, + // keep_alive: $settings.keepAlive ?? undefined, + // stop: + // (params?.stop ?? $settings?.params?.stop ?? undefined) + // ? ( + // params?.stop.split(',').map((token) => token.trim()) ?? $settings.params.stop + // ).map((str) => + // decodeURIComponent(JSON.parse('"' + str.replace(/\"/g, '\\"') + '"')) + // ) + // : undefined + // }, + tool_ids: selectedToolIds.length > 0 ? selectedToolIds : undefined, files: files.length > 0 ? files : undefined, session_id: $socket?.id, chat_id: $chatId, - id: responseMessageId + id: responseMessageId, + + ...(stream && (model.info?.meta?.capabilities?.usage ?? false) + ? { + stream_options: { + include_usage: true + } + } + : {}) }, `${WEBUI_BASE_URL}/api` ); @@ -1636,6 +1637,7 @@ await handleOpenAIError(error, null, model, responseMessage); break; } + if (done || stopResponseFlag || _chatId !== $chatId) { responseMessage.done = true; history.messages[responseMessageId] = responseMessage; @@ -1648,7 +1650,7 @@ } if (usage) { - responseMessage.info = { ...usage, openai: true, usage }; + responseMessage.usage = usage; } if (selectedModelId) { diff --git a/src/lib/components/chat/Messages/CitationsModal.svelte b/src/lib/components/chat/Messages/CitationsModal.svelte index 9c6b23ecd..e81daada9 100644 --- a/src/lib/components/chat/Messages/CitationsModal.svelte +++ b/src/lib/components/chat/Messages/CitationsModal.svelte @@ -2,6 +2,7 @@ import { getContext, onMount, tick } from 'svelte'; import Modal from '$lib/components/common/Modal.svelte'; import Tooltip from '$lib/components/common/Tooltip.svelte'; + import { WEBUI_API_BASE_URL } from '$lib/constants'; const i18n = getContext('i18n'); @@ -91,7 +92,7 @@ {/if} - {#if message.info} + {#if message.usage} ${sanitizeResponseContent( - JSON.stringify(message.info.usage, null, 2) - .replace(/"([^(")"]+)":/g, '$1:') - .slice(1, -1) - .split('\n') - .map((line) => line.slice(2)) - .map((line) => (line.endsWith(',') ? line.slice(0, -1) : line)) - .join('\n') - )}` - : `prompt_tokens: ${message.info.prompt_tokens ?? 'N/A'}
- completion_tokens: ${message.info.completion_tokens ?? 'N/A'}
- total_tokens: ${message.info.total_tokens ?? 'N/A'}` - : `response_token/s: ${ - `${ - Math.round( - ((message.info.eval_count ?? 0) / - ((message.info.eval_duration ?? 0) / 1000000000)) * - 100 - ) / 100 - } tokens` ?? 'N/A' - }
- prompt_token/s: ${ - Math.round( - ((message.info.prompt_eval_count ?? 0) / - ((message.info.prompt_eval_duration ?? 0) / 1000000000)) * - 100 - ) / 100 ?? 'N/A' - } tokens
- total_duration: ${ - Math.round(((message.info.total_duration ?? 0) / 1000000) * 100) / 100 ?? 'N/A' - }ms
- load_duration: ${ - Math.round(((message.info.load_duration ?? 0) / 1000000) * 100) / 100 ?? 'N/A' - }ms
- prompt_eval_count: ${message.info.prompt_eval_count ?? 'N/A'}
- prompt_eval_duration: ${ - Math.round(((message.info.prompt_eval_duration ?? 0) / 1000000) * 100) / 100 ?? - 'N/A' - }ms
- eval_count: ${message.info.eval_count ?? 'N/A'}
- eval_duration: ${ - Math.round(((message.info.eval_duration ?? 0) / 1000000) * 100) / 100 ?? 'N/A' - }ms
- approximate_total: ${approximateToHumanReadable(message.info.total_duration ?? 0)}`} - placement="top" + content={message.usage + ? `
${sanitizeResponseContent(
+													JSON.stringify(message.usage, null, 2)
+														.replace(/"([^(")"]+)":/g, '$1:')
+														.slice(1, -1)
+														.split('\n')
+														.map((line) => line.slice(2))
+														.map((line) => (line.endsWith(',') ? line.slice(0, -1) : line))
+														.join('\n')
+												)}
` + : ''} + placement="bottom" > - - - + + + {/if} diff --git a/src/lib/components/common/CodeEditor.svelte b/src/lib/components/common/CodeEditor.svelte index b521978f4..b9ecfb239 100644 --- a/src/lib/components/common/CodeEditor.svelte +++ b/src/lib/components/common/CodeEditor.svelte @@ -6,7 +6,7 @@ import { acceptCompletion } from '@codemirror/autocomplete'; import { indentWithTab } from '@codemirror/commands'; - import { indentUnit } from '@codemirror/language'; + import { indentUnit, LanguageDescription } from '@codemirror/language'; import { languages } from '@codemirror/language-data'; import { oneDark } from '@codemirror/theme-one-dark'; @@ -47,6 +47,15 @@ let editorTheme = new Compartment(); let editorLanguage = new Compartment(); + languages.push( + LanguageDescription.of({ + name: 'HCL', + extensions: ['hcl', 'tf'], + load() { + return import('codemirror-lang-hcl').then((m) => m.hcl()); + } + }) + ); const getLang = async () => { const language = languages.find((l) => l.alias.includes(lang)); return await language?.load(); diff --git a/src/lib/constants.ts b/src/lib/constants.ts index 700bd3c42..d92f33671 100644 --- a/src/lib/constants.ts +++ b/src/lib/constants.ts @@ -9,9 +9,9 @@ export const WEBUI_API_BASE_URL = `${WEBUI_BASE_URL}/api/v1`; export const OLLAMA_API_BASE_URL = `${WEBUI_BASE_URL}/ollama`; export const OPENAI_API_BASE_URL = `${WEBUI_BASE_URL}/openai`; -export const AUDIO_API_BASE_URL = `${WEBUI_BASE_URL}/audio/api/v1`; -export const IMAGES_API_BASE_URL = `${WEBUI_BASE_URL}/images/api/v1`; -export const RETRIEVAL_API_BASE_URL = `${WEBUI_BASE_URL}/retrieval/api/v1`; +export const AUDIO_API_BASE_URL = `${WEBUI_BASE_URL}/api/v1/audio`; +export const IMAGES_API_BASE_URL = `${WEBUI_BASE_URL}/api/v1/images`; +export const RETRIEVAL_API_BASE_URL = `${WEBUI_BASE_URL}/api/v1/retrieval`; export const WEBUI_VERSION = APP_VERSION; export const WEBUI_BUILD_HASH = APP_BUILD_HASH; diff --git a/src/lib/i18n/locales/cs-CZ/translation.json b/src/lib/i18n/locales/cs-CZ/translation.json index 2ec6671c9..90623afa8 100644 --- a/src/lib/i18n/locales/cs-CZ/translation.json +++ b/src/lib/i18n/locales/cs-CZ/translation.json @@ -1,6 +1,6 @@ { "-1 for no limit, or a positive integer for a specific limit": "", - "'s', 'm', 'h', 'd', 'w' or '-1' for no expiration.": "'s', 'm', 'h', 'd', 'w' nebo '-1' pro žádné vypršení platnosti.", + "'s', 'm', 'h', 'd', 'w' or '-1' for no expiration.": "'s', 'm', 'h', 'd', 'w' nebo '-1' pro žádné vypršení.", "(e.g. `sh webui.sh --api --api-auth username_password`)": "(např. `sh webui.sh --api --api-auth username_password`)", "(e.g. `sh webui.sh --api`)": "(např. `sh webui.sh --api`)", "(latest)": "Nejnovější", @@ -493,7 +493,7 @@ "JSON Preview": "Náhled JSON", "July": "Červenec", "June": "červen", - "JWT Expiration": "Vypršení platnosti JWT (JSON Web Token)", + "JWT Expiration": "Vypršení JWT", "JWT Token": "JWT Token (JSON Web Token)", "Keep Alive": "Udržovat spojení", "Key": "", @@ -659,8 +659,8 @@ "Permission denied when accessing microphone: {{error}}": "Oprávnění zamítnuto při přístupu k mikrofonu: {{error}}", "Permissions": "", "Personalization": "Personalizace", - "Pin": "Kolík", - "Pinned": "Připnuto", + "Pin": "", + "Pinned": "", "Pioneer insights": "", "Pipeline deleted successfully": "Pipeline byla úspěšně odstraněna", "Pipeline downloaded successfully": "Kanál byl úspěšně stažen", diff --git a/src/lib/i18n/locales/languages.json b/src/lib/i18n/locales/languages.json index 809d375e4..5672e4592 100644 --- a/src/lib/i18n/locales/languages.json +++ b/src/lib/i18n/locales/languages.json @@ -147,6 +147,10 @@ "code": "ru-RU", "title": "Russian (Russia)" }, + { + "code": "sk-SK", + "title": "Slovak (Slovenčina)" + }, { "code": "sv-SE", "title": "Swedish (Svenska)" diff --git a/src/lib/i18n/locales/nb-NO/translation.json b/src/lib/i18n/locales/nb-NO/translation.json index 4c8a3397f..93ed4449f 100644 --- a/src/lib/i18n/locales/nb-NO/translation.json +++ b/src/lib/i18n/locales/nb-NO/translation.json @@ -12,14 +12,14 @@ "A task model is used when performing tasks such as generating titles for chats and web search queries": "En oppgavemodell brukes når du utfører oppgaver som å generere titler for samtaler eller utfører søkeforespørsler på nettet", "a user": "en bruker", "About": "Om", - "Access": "", - "Access Control": "", - "Accessible to all users": "", + "Access": "Tilgang", + "Access Control": "Tilgangskontroll", + "Accessible to all users": "Tilgjengelig for alle brukere", "Account": "Konto", "Account Activation Pending": "Venter på kontoaktivering", "Accurate information": "Nøyaktig informasjon", "Actions": "Handlinger", - "Activate this command by typing \"/{{COMMAND}}\" to chat input.": "Aktiver denne kommandoen ved å skrive inn \"/{{COMMAND}}\" i chattens inntastingsfelt", + "Activate this command by typing \"/{{COMMAND}}\" to chat input.": "Aktiver denne kommandoen ved å skrive inn \"/{{COMMAND}}\" i chattens inntastingsfelt.", "Active Users": "Aktive brukere", "Add": "Legg til", "Add a model ID": "Legg til en modell-ID", @@ -29,38 +29,38 @@ "Add Connection": "Legg til tilkobling", "Add Content": "Legg til innhold", "Add content here": "Legg til innhold her", - "Add custom prompt": "Legg til egendefinert prompt", + "Add custom prompt": "Legg til tilpasset ledetekst", "Add Files": "Legg til filer", - "Add Group": "", + "Add Group": "Legg til gruppe", "Add Memory": "Legg til minne", "Add Model": "Legg til modell", "Add Tag": "Legg til etikett", "Add Tags": "Legg til etiketter", "Add text content": "Legg til tekstinnhold", "Add User": "Legg til bruker", - "Add User Group": "", + "Add User Group": "Legg til brukergruppe", "Adjusting these settings will apply changes universally to all users.": "Endring av disse innstillingene vil gjelde for alle brukere på tvers av systemet.", "admin": "administrator", "Admin": "Administrator", "Admin Panel": "Administratorpanel", "Admin Settings": "Administratorinnstillinger", - "Admins have access to all tools at all times; users need tools assigned per model in the workspace.": "Administratorer har alltid tilgang til alle verktøy. Brukere må få tildelt verktøy for hver enkelt modell i arbeidsområdet.", + "Admins have access to all tools at all times; users need tools assigned per model in the workspace.": "Administratorer har alltid tilgang til alle verktøy. Brukere må få tildelt verktøy per modell i arbeidsområdet.", "Advanced Parameters": "Avanserte parametere", "Advanced Params": "Avanserte parametere", "All chats": "Alle chatter", "All Documents": "Alle dokumenter", - "All models deleted successfully": "", - "Allow Chat Delete": "", + "All models deleted successfully": "Alle modeller er slettet", + "Allow Chat Delete": "Tillat sletting av chatter", "Allow Chat Deletion": "Tillat sletting av chatter", - "Allow Chat Edit": "", - "Allow File Upload": "", + "Allow Chat Edit": "Tillat redigering av chatter", + "Allow File Upload": "Tillatt opplasting av filer", "Allow non-local voices": "Tillat ikke-lokale stemmer", "Allow Temporary Chat": "Tillat midlertidige chatter", "Allow User Location": "Aktiver stedstjenester", - "Allow Voice Interruption in Call": "Muliggjør stemmeavbrytelse i samtaler", + "Allow Voice Interruption in Call": "Muliggjør taleavbrytelse i samtaler", "Already have an account?": "Har du allerede en konto?", "Alternative to the top_p, and aims to ensure a balance of quality and variety. The parameter p represents the minimum probability for a token to be considered, relative to the probability of the most likely token. For example, with p=0.05 and the most likely token having a probability of 0.9, logits with a value less than 0.045 are filtered out. (Default: 0.0)": "Alternativ til top_p, og har som mål å sikre en balanse mellom kvalitet og variasjon. Parameteren p representerer minimumssannsynligheten for at et token skal vurderes, i forhold til sannsynligheten for det mest sannsynlige tokenet. Hvis p for eksempel er 0,05 og det mest sannsynlige tokenet har en sannsynlighet på 0,9, filtreres logits med en verdi på mindre enn 0,045 bort. (Standard: 0,0)", - "Amazing": "", + "Amazing": "Flott", "an assistant": "en assistent", "and": "og", "and {{COUNT}} more": "og {{COUNT}} til", @@ -71,7 +71,7 @@ "API keys": "API-nøkler", "Application DN": "Applikasjonens DN", "Application DN Password": "Applikasjonens DN-passord", - "applies to all users with the \"user\" role": "", + "applies to all users with the \"user\" role": "gjelder for alle brukere med rollen \"user\"", "April": "april", "Archive": "Arkiv", "Archive All Chats": "Arkiver alle chatter", @@ -84,35 +84,35 @@ "Ask a question": "Still et spørsmål", "Assistant": "Assistent", "Attach file": "Legg ved fil", - "Attention to detail": "Sans for detaljer", + "Attention to detail": "Fokus på detaljer", "Attribute for Username": "Attributt for brukernavn", "Audio": "Lyd", "August": "august", "Authenticate": "Godkjenn", - "Auto-Copy Response to Clipboard": "Respons auto-kopi til utklippstavle", - "Auto-playback response": "Automatisk avspilling av svar", + "Auto-Copy Response to Clipboard": "Kopier svar automatisk til utklippstavlen", + "Auto-playback response": "Spill av svar automatisk", "Autocomplete Generation": "", "Autocomplete Generation Input Max Length": "", "Automatic1111": "Automatic1111", - "AUTOMATIC1111 Api Auth String": "AUTOMATIC1111 Api Autentiseringsstreng", - "AUTOMATIC1111 Base URL": "AUTOMATIC1111 Grunn-URL", - "AUTOMATIC1111 Base URL is required.": "AUTOMATIC1111 Grunn-URL kreves.", + "AUTOMATIC1111 Api Auth String": "API-Autentiseringsstreng for AUTOMATIC1111", + "AUTOMATIC1111 Base URL": "Absolutt URL for AUTOMATIC1111", + "AUTOMATIC1111 Base URL is required.": "Absolutt URL for AUTOMATIC1111 kreves.", "Available list": "Tilgjengelig liste", "available!": "tilgjengelig!", - "Awful": "", + "Awful": "Fælt", "Azure AI Speech": "Azure AI-tale", "Azure Region": "Azure område", "Back": "Tilbake", "Bad Response": "Dårlig svar", "Banners": "Bannere", - "Base Model (From)": "Grunnmodell (Fra)", + "Base Model (From)": "Grunnmodell (fra)", "Batch Size (num_batch)": "Batchstørrelse (num_batch)", "before": "før", "Being lazy": "Er lat", "Bing Search V7 Endpoint": "Endepunkt for Bing Search V7", "Bing Search V7 Subscription Key": "Abonnementsnøkkel for Bing Search V7", "Brave Search API Key": "API-nøkkel for Brave Search", - "By {{name}}": "", + "By {{name}}": "Etter {{name}}", "Bypass SSL verification for Websites": "Omgå SSL-verifisering for nettsteder", "Call": "Ring", "Call feature is not supported when using Web STT engine": "Ringefunksjonen støttes ikke når du bruker Web STT-motoren", @@ -130,7 +130,7 @@ "Chat Controls": "Kontrollere i chat", "Chat direction": "Retning på chat", "Chat Overview": "Chatoversikt", - "Chat Permissions": "", + "Chat Permissions": "Tillatelser for chat", "Chat Tags Auto-Generation": "Auto-generering av chatetiketter", "Chats": "Chatter", "Check Again": "Sjekk på nytt", @@ -141,7 +141,7 @@ "Chunk Params": "Chunk-parametere", "Chunk Size": "Chunk-størrelse", "Ciphers": "Chiffer", - "Citation": "Sitering", + "Citation": "Kildehenvisning", "Clear memory": "Tøm minnet", "click here": "Klikk her", "Click here for filter guides.": "Klikk her for å få veiledning om filtre", @@ -152,7 +152,7 @@ "Click here to select": "Klikk her for å velge", "Click here to select a csv file.": "Klikk her for å velge en CSV-fil.", "Click here to select a py file.": "Klikk her for å velge en PY-fil.", - "Click here to upload a workflow.json file.": "Klikk her for å laste opp en workflow.json fil.", + "Click here to upload a workflow.json file.": "Klikk her for å laste opp en workflow.json-fil.", "click here.": "klikk her.", "Click on the user role button to change a user's role.": "Klikk på knappen Brukerrolle for å endre en brukers rolle.", "Clipboard write permission denied. Please check your browser settings to grant the necessary access.": "Skrivetilgang til utklippstavlen avslått. Kontroller nettleserinnstillingene for å gi den nødvendige tilgangen.", @@ -161,7 +161,7 @@ "Code execution": "Kodekjøring", "Code formatted successfully": "Koden er formatert", "Collection": "Samling", - "Color": "", + "Color": "Farge", "ComfyUI": "ComfyUI", "ComfyUI Base URL": "Absolutt URL for ComfyUI", "ComfyUI Base URL is required.": "Absolutt URL for ComfyUI kreves.", @@ -171,7 +171,7 @@ "Completions": "Fullføringer", "Concurrent Requests": "Samtidige forespørsler", "Configure": "Konfigurer", - "Configure Models": "", + "Configure Models": "Konfigurer modeller", "Confirm": "Bekreft", "Confirm Password": "Bekreft passordet", "Confirm your action": "Bekreft handlingen", @@ -196,12 +196,12 @@ "Copy Link": "Kopier lenke", "Copy to clipboard": "Kopier til utklippstavle", "Copying to clipboard was successful!": "Kopiert til utklippstavlen!", - "Create": "", + "Create": "Opprett", "Create a knowledge base": "Opprett en kunnskapsbase", "Create a model": "Opprett en modell", "Create Account": "Opprett konto", "Create Admin Account": "Opprett administratorkonto", - "Create Group": "", + "Create Group": "Opprett gruppe", "Create Knowledge": "Opprett kunnskap", "Create new key": "Lag ny nøkkel", "Create new secret key": "Lag ny hemmelig nøkkel", @@ -220,17 +220,17 @@ "Default (SentenceTransformers)": "Standard (SentenceTransformers)", "Default Model": "Standard modell", "Default model updated": "Standard modell oppdatert", - "Default Models": "", - "Default permissions": "", - "Default permissions updated successfully": "", + "Default Models": "Standard modeller", + "Default permissions": "Standard tillatelser", + "Default permissions updated successfully": "Standard tillatelser oppdatert", "Default Prompt Suggestions": "Standard forslag til ledetekster", - "Default to 389 or 636 if TLS is enabled": "Velg 389 or 636 som standard hvis TLS er aktivert", + "Default to 389 or 636 if TLS is enabled": "Velg 389 eller 636 som standard hvis TLS er aktivert", "Default to ALL": "Velg ALL som standard", "Default User Role": "Standard brukerrolle", "Delete": "Slett", "Delete a model": "Slett en modell", "Delete All Chats": "Slett alle chatter", - "Delete All Models": "", + "Delete All Models": "Slett alle modeller", "Delete chat": "Slett chat", "Delete Chat": "Slett chat", "Delete chat?": "Slette chat?", @@ -242,7 +242,7 @@ "Delete User": "Slett bruker", "Deleted {{deleteModelTag}}": "Slettet {{deleteModelTag}}", "Deleted {{name}}": "Slettet {{name}}", - "Deleted User": "", + "Deleted User": "Slettet bruker", "Describe your knowledge base and objectives": "Beskriv kunnskapsbasen din og målene dine", "Description": "Beskrivelse", "Didn't fully follow instructions": "Fulgte ikke instruksjonene fullstendig", @@ -257,10 +257,10 @@ "Discover, download, and explore custom tools": "Oppdag, last ned og utforsk tilpassede verktøy", "Discover, download, and explore model presets": "Oppdag, last ned og utforsk forhåndsinnstillinger for modeller", "Dismissible": "Kan lukkes", - "Display": "", + "Display": "Visning", "Display Emoji in Call": "Vis emoji i samtale", - "Display the username instead of You in the Chat": "Vis brukernavnet i stedet for Du i chatten", - "Displays citations in the response": "", + "Display the username instead of You in the Chat": "Vis brukernavnet ditt i stedet for Du i chatten", + "Displays citations in the response": "Vis kildehenvisninger i svaret", "Dive into knowledge": "Bli kjent med kunnskap", "Do not install functions from sources you do not fully trust.": "Ikke installer funksjoner fra kilder du ikke stoler på.", "Do not install tools from sources you do not fully trust.": "Ikke installer verktøy fra kilder du ikke stoler på.", @@ -276,23 +276,23 @@ "Download": "Last ned", "Download canceled": "Nedlasting avbrutt", "Download Database": "Last ned database", - "Drag and drop a file to upload or select a file to view": "", + "Drag and drop a file to upload or select a file to view": "Dra og slipp en fil for å laste den opp, eller velg en fil å vise den", "Draw": "Tegne", "Drop any files here to add to the conversation": "Slipp filer her for å legge dem til i samtalen", "e.g. '30s','10m'. Valid time units are 's', 'm', 'h'.": "f.eks. '30s','10m'. Gyldige tidsenheter er 's', 'm', 't'.", "e.g. A filter to remove profanity from text": "f.eks. et filter for å fjerne banning fra tekst", "e.g. My Filter": "f.eks. Mitt filter", - "e.g. My Tools": "", + "e.g. My Tools": "f.eks. Mine verktøy", "e.g. my_filter": "f.eks. mitt_filter", - "e.g. my_tools": "", - "e.g. Tools for performing various operations": "", + "e.g. my_tools": "f.eks. mine_verktøy", + "e.g. Tools for performing various operations": "f.eks. Verktøy for å gjøre ulike handlinger", "Edit": "Rediger", "Edit Arena Model": "Rediger Arena-modell", "Edit Connection": "Rediger tilkobling", - "Edit Default Permissions": "", + "Edit Default Permissions": "Rediger standard tillatelser", "Edit Memory": "Rediger minne", "Edit User": "Rediger bruker", - "Edit User Group": "", + "Edit User Group": "Rediger brukergruppe", "ElevenLabs": "ElevenLabs", "Email": "E-postadresse", "Embark on adventures": "Kom med på eventyr", @@ -300,8 +300,8 @@ "Embedding Model": "Innbyggingsmodell", "Embedding Model Engine": "Motor for innbygging av modeller", "Embedding model set to \"{{embedding_model}}\"": "Innbyggingsmodell angitt til \"{{embedding_model}}\"", - "Enable API Key Auth": "", - "Enable autocomplete generation for chat messages": "", + "Enable API Key Auth": "Aktiver godkjenning med API-nøkkel", + "Enable autocomplete generation for chat messages": "Aktiver automatisk utfylling av chatmeldinger", "Enable Community Sharing": "Aktiver deling i fellesskap", "Enable Memory Locking (mlock) to prevent model data from being swapped out of RAM. This option locks the model's working set of pages into RAM, ensuring that they will not be swapped out to disk. This can help maintain performance by avoiding page faults and ensuring fast data access.": "Aktiver Memory Locking (mlock) for å forhindre at modelldata byttes ut av RAM. Dette alternativet låser modellens arbeidssett med sider i RAM-minnet, slik at de ikke byttes ut til disk. Dette kan bidra til å opprettholde ytelsen ved å unngå sidefeil og sikre rask datatilgang.", "Enable Memory Mapping (mmap) to load model data. This option allows the system to use disk storage as an extension of RAM by treating disk files as if they were in RAM. This can improve model performance by allowing for faster data access. However, it may not work correctly with all systems and can consume a significant amount of disk space.": "Aktiver Memory Mapping (mmap) for å laste inn modelldata. Med dette alternativet kan systemet bruke disklagring som en utvidelse av RAM ved å behandle diskfiler som om de befant seg i RAM. Dette kan forbedre modellens ytelse ved å gi raskere datatilgang. Det er imidlertid ikke sikkert at det fungerer som det skal på alle systemer, og det kan kreve mye diskplass.", @@ -311,10 +311,10 @@ "Enable Web Search": "Aktiver websøk", "Enabled": "Aktivert", "Engine": "Motor", - "Ensure your CSV file includes 4 columns in this order: Name, Email, Password, Role.": "Sørg for at CSV-filen din inkluderer 4 kolonner i denne rekkefølgen: Navn, E-post, Passord, Rolle.", + "Ensure your CSV file includes 4 columns in this order: Name, Email, Password, Role.": "Sørg for at CSV-filen din inkluderer fire kolonner i denne rekkefølgen: Navn, E-post, Passord, Rolle.", "Enter {{role}} message here": "Skriv inn {{role}} melding her", "Enter a detail about yourself for your LLMs to recall": "Skriv inn en detalj om deg selv som språkmodellene dine kan huske", - "Enter api auth string (e.g. username:password)": "Skriv inn api-autentiseringsstreng (f.eks. brukernavn:passord)", + "Enter api auth string (e.g. username:password)": "Skriv inn API-autentiseringsstreng (f.eks. brukernavn:passord)", "Enter Application DN": "Angi applikasjonens DN", "Enter Application DN Password": "Angi applikasjonens DN-passord", "Enter Bing Search V7 Endpoint": "Angi endepunkt for Bing Search V7", @@ -333,7 +333,7 @@ "Enter language codes": "Angi språkkoder", "Enter Model ID": "Angi modellens ID", "Enter model tag (e.g. {{modelTag}})": "Angi modellens etikett (f.eks. {{modelTag}})", - "Enter Mojeek Search API Key": "", + "Enter Mojeek Search API Key": "Angi API-nøkkel for Mojeek-søk", "Enter Number of Steps (e.g. 50)": "Angi antall steg (f.eks. 50)", "Enter proxy URL (e.g. https://user:password@host:port)": "", "Enter Sampler (e.g. Euler a)": "Angi Sampler (e.g. Euler a)", @@ -358,7 +358,7 @@ "Enter URL (e.g. http://localhost:11434)": "Angi URL (f.eks. http://localhost:11434)", "Enter Your Email": "Skriv inn e-postadressen din", "Enter Your Full Name": "Skriv inn det fulle navnet ditt", - "Enter your message": "Skriv inn meldingen din", + "Enter your message": "Skriv inn din melding", "Enter Your Password": "Skriv inn passordet ditt", "Enter Your Role": "Skriv inn rollen din", "Enter Your Username": "Skriv inn brukernavnet ditt", @@ -380,19 +380,19 @@ "Export Config to JSON File": "Ekporter konfigurasjon til en JSON-fil", "Export Functions": "Eksporter funksjoner", "Export Models": "Eksporter modeller", - "Export Presets": "", + "Export Presets": "Eksporter forhåndsinnstillinger", "Export Prompts": "Eksporter ledetekster", "Export to CSV": "Eksporter til CSV", "Export Tools": "Eksporter verktøy", "External Models": "Eksterne modeller", "Failed to add file.": "Kan ikke legge til filen.", "Failed to create API Key.": "Kan ikke opprette en API-nøkkel.", - "Failed to read clipboard contents": "Kan ikke lese innhold på utklippstavlen", - "Failed to save models configuration": "", + "Failed to read clipboard contents": "Kan ikke lese utklippstavlens innhold", + "Failed to save models configuration": "Kan ikke lagre konfigurasjonen av modeller", "Failed to update settings": "Kan ikke oppdatere innstillinger", "Failed to upload file.": "Kan ikke laste opp filen.", "February": "februar", - "Feedback History": "Tilbakemeldingshistorikk", + "Feedback History": "Tilbakemeldingslogg", "Feedbacks": "Tilbakemeldinger", "Feel free to add specific details": "Legg gjerne til bestemte detaljer", "File": "Fil", @@ -408,7 +408,7 @@ "Filters": "Filtre", "Fingerprint spoofing detected: Unable to use initials as avatar. Defaulting to default profile image.": "Fingeravtrykk-spoofing oppdaget: kan ikke bruke initialer som avatar. Bruker standard profilbilde.", "Fluidly stream large external response chunks": "Flytende strømming av store eksterne svarpakker", - "Focus chat input": "Fokuser på chat-inndata", + "Focus chat input": "Fokusert chat-inndata", "Folder deleted successfully": "Mappe slettet", "Folder name cannot be empty": "Mappenavn kan ikke være tomt", "Folder name cannot be empty.": "Mappenavn kan ikke være tomt.", @@ -442,11 +442,11 @@ "Good Response": "Godt svar", "Google PSE API Key": "API-nøkkel for Google PSE", "Google PSE Engine Id": "Motor-ID for Google PSE", - "Group created successfully": "", - "Group deleted successfully": "", - "Group Description": "", - "Group Name": "", - "Group updated successfully": "", + "Group created successfully": "Gruppe opprettet", + "Group deleted successfully": "Gruppe slettet", + "Group Description": "Beskrivelse av gruppe", + "Group Name": "Navn på gruppe", + "Group updated successfully": "Gruppe oppdatert", "Groups": "Grupper", "h:mm a": "t:mm a", "Haptic Feedback": "Haptisk tilbakemelding", @@ -454,12 +454,12 @@ "Hello, {{name}}": "Hei, {{name}}!", "Help": "Hjelp", "Help us create the best community leaderboard by sharing your feedback history!": "Hjelp oss med å skape den beste fellesskapsledertavlen ved å dele tilbakemeldingshistorikken din.", - "Hex Color": "", - "Hex Color - Leave empty for default color": "", + "Hex Color": "Hex-farge", + "Hex Color - Leave empty for default color": "Hex-farge – la stå tom for standard farge", "Hide": "Skjul", "Host": "Host", "How can I help you today?": "Hva kan jeg hjelpe deg med i dag?", - "How would you rate this response?": "", + "How would you rate this response?": "Hvordan vurderer du dette svaret?", "Hybrid Search": "Hybrid-søk", "I acknowledge that I have read and I understand the implications of my action. I am aware of the risks associated with executing arbitrary code and I have verified the trustworthiness of the source.": "Jeg bekrefter at jeg har lest og forstår konsekvensene av mine handlinger. Jeg er klar over risikoen forbundet med å kjøre vilkårlig kode, og jeg har verifisert kildens pålitelighet.", "ID": "ID", @@ -472,7 +472,7 @@ "Import Config from JSON File": "Importer konfigurasjon fra en JSON-fil", "Import Functions": "Importer funksjoner", "Import Models": "Importer modeller", - "Import Presets": "", + "Import Presets": "Importer forhåndsinnstillinger", "Import Prompts": "Importer ledetekster", "Import Tools": "Importer verktøy", "Include": "Inkluder", @@ -482,7 +482,7 @@ "Info": "Info", "Input commands": "Inntast kommandoer", "Install from Github URL": "Installer fra GitHub-URL", - "Instant Auto-Send After Voice Transcription": "Øyeblikkelig automatisk sending etter stemmetranskripsjon", + "Instant Auto-Send After Voice Transcription": "Øyeblikkelig automatisk sending etter taletranskripsjon", "Interface": "Grensesnitt", "Invalid file format.": "Ugyldig filformat.", "Invalid Tag": "Ugyldig etikett", @@ -499,7 +499,7 @@ "Key": "Nøkkel", "Keyboard shortcuts": "Hurtigtaster", "Knowledge": "Kunnskap", - "Knowledge Access": "", + "Knowledge Access": "Tilgang til kunnskap", "Knowledge created successfully.": "Kunnskap opprettet.", "Knowledge deleted successfully.": "Kunnskap slettet.", "Knowledge reset successfully.": "Tilbakestilling av kunnskap vellykket.", @@ -529,7 +529,7 @@ "Make sure to export a workflow.json file as API format from ComfyUI.": "Sørg for å eksportere en workflow.json-fil i API-formatet fra ComfyUI.", "Manage": "Administrer", "Manage Arena Models": "Behandle Arena-modeller", - "Manage Ollama": "", + "Manage Ollama": "Behandle Ollama", "Manage Ollama API Connections": "Behandle API-tilkoblinger for Ollama", "Manage OpenAI API Connections": "Behandle API-tilkoblinger for OpenAPI", "Manage Pipelines": "Behandle pipelines", @@ -546,7 +546,7 @@ "Memory deleted successfully": "Minne slettet", "Memory updated successfully": "Minne oppdatert", "Merge Responses": "Flette svar", - "Message rating should be enabled to use this feature": "Vurdering av meldinger må være aktivert for å kunne bruke denne funksjonen", + "Message rating should be enabled to use this feature": "Vurdering av meldinger må være aktivert for å ta i bruk denne funksjonen", "Messages you send after creating your link won't be shared. Users with the URL will be able to view the shared chat.": "Meldinger du sender etter at du har opprettet lenken, blir ikke delt. Brukere med URL-en vil kunne se den delte chatten.", "Min P": "Min P", "Minimum Score": "Minimum poengsum", @@ -557,27 +557,25 @@ "MMMM DD, YYYY HH:mm": "HH:mm DD MMMM YYYY", "MMMM DD, YYYY hh:mm:ss A": "hh:mm:ss A DD MMMM YYYY", "Model": "Modell", - "Model '{{modelName}}' has been successfully downloaded.": "Modellen '{{modelName}}' er lastet ned.", - "Model '{{modelTag}}' is already in queue for downloading.": "Modellen '{{modelTag}}' er allerede i nedlastingskøen.", + "Model '{{modelName}}' has been successfully downloaded.": "Modellen {{modelName}} er lastet ned.", + "Model '{{modelTag}}' is already in queue for downloading.": "Modellen {{modelTag}} er allerede i nedlastingskøen.", "Model {{modelId}} not found": "Finner ikke modellen {{modelId}}", "Model {{modelName}} is not vision capable": "Modellen {{modelName}} er ikke egnet til visuelle data", "Model {{name}} is now {{status}}": "Modellen {{name}} er nå {{status}}", "Model accepts image inputs": "Modellen godtar bildeinndata", "Model created successfully!": "Modellen er opprettet!", "Model filesystem path detected. Model shortname is required for update, cannot continue.": "Modellfilsystembane oppdaget. Kan ikke fortsette fordi modellens kortnavn er påkrevd for oppdatering.", - "Model Filtering": "", + "Model Filtering": "Filtrering av modeller", "Model ID": "Modell-ID", "Model IDs": "Modell-ID-er", "Model Name": "Modell", "Model not selected": "Modell ikke valgt", "Model Params": "Modellparametere", - "Model Permissions": "", + "Model Permissions": "Modelltillatelser", "Model updated successfully": "Modell oppdatert", "Modelfile Content": "Modellfilinnhold", "Models": "Modeller", - "Models Access": "", - "Models configuration saved successfully": "", - "Mojeek Search API Key": "", + "Models Access": "Tilgang til modeller", "more": "mer", "More": "Mer", "Name": "Navn", @@ -591,16 +589,16 @@ "No feedbacks found": "Finner ingen tilbakemeldinger", "No file selected": "Ingen fil valgt", "No files found.": "Finner ingen filer", - "No groups with access, add a group to grant access": "", + "No groups with access, add a group to grant access": "Ingen grupper med tilgang. Legg til en gruppe som skal ha tilgang.", "No HTML, CSS, or JavaScript content found.": "Finner ikke noe HTML, CSS- eller JavaScript-innhold.", "No knowledge found": "Finner ingen kunnskaper", - "No model IDs": "", + "No model IDs": "Ingen modell-ID-er", "No models found": "Finner ingen modeller", - "No models selected": "", + "No models selected": "Ingen modeller er valgt", "No results found": "Finner ingen resultater", "No search query generated": "Ingen søkespørringer er generert", "No source available": "Ingen kilde tilgjengelig", - "No users were found.": "", + "No users were found.": "Finner ingen brukere", "No valves to update": "Ingen ventiler å oppdatere", "None": "Ingen", "Not factually correct": "Uriktig informasjon", @@ -617,16 +615,16 @@ "Okay, Let's Go!": "OK, kjør på!", "OLED Dark": "OLED mørk", "Ollama": "Ollama", - "Ollama API": "Ollama API", - "Ollama API disabled": "Ollama API deaktivert", + "Ollama API": "Ollama-API", + "Ollama API disabled": "Ollama-API deaktivert", "Ollama API settings updated": "API-innstillinger for Ollama er oppdatert", "Ollama Version": "Ollama-versjon", "On": "Aktivert", "Only alphanumeric characters and hyphens are allowed": "Bare alfanumeriske tegn og bindestreker er tillatt", "Only alphanumeric characters and hyphens are allowed in the command string.": "Bare alfanumeriske tegn og bindestreker er tillatt i kommandostrengen.", "Only collections can be edited, create a new knowledge base to edit/add documents.": "Bare samlinger kan redigeres, eller lag en ny kunnskapsbase for å kunne redigere / legge til dokumenter.", - "Only select users and groups with permission can access": "", - "Oops! Looks like the URL is invalid. Please double-check and try again.": "Oi! Det ser ut som URL-en er ugyldig. Dobbeltsjekk, og prøv igjen.", + "Only select users and groups with permission can access": "Bare utvalgte brukere og grupper med tillatelse kan få tilgang", + "Oops! Looks like the URL is invalid. Please double-check and try again.": "Oi! Det ser ut som URL-en er ugyldig. Dobbeltsjekk, og prøv på nytt.", "Oops! There are files still uploading. Please wait for the upload to complete.": "Oi! Det er fortsatt filer som lastes opp. Vent til opplastingen er ferdig.", "Oops! There was an error in the previous response.": "Oi! Det er en feil i det forrige svaret.", "Oops! You're using an unsupported method (frontend only). Please serve the WebUI from the backend.": "Oi! Du bruker en ikke-støttet metode (bare frontend). Du må kjøre WebUI fra backend.", @@ -637,27 +635,27 @@ "Open WebUI uses SpeechT5 and CMU Arctic speaker embeddings.": "Open WebUI bruker SpeechT5 og CMU Arctic-høytalerinnbygginger", "Open WebUI version (v{{OPEN_WEBUI_VERSION}}) is lower than required version (v{{REQUIRED_VERSION}})": "Open WebUI-versjonen (v{{OPEN_WEBUI_VERSION}}) er lavere enn den påkrevde versjonen (v{{REQUIRED_VERSION}})", "OpenAI": "OpenAI", - "OpenAI API": "OpenAI API", + "OpenAI API": "OpenAI-API", "OpenAI API Config": "API-konfigurasjon for OpenAI", "OpenAI API Key is required.": "API-nøkkel for OpenAI kreves.", "OpenAI API settings updated": "API-innstillinger for OpenAI er oppdatert", "OpenAI URL/Key required.": "URL/nøkkel for OpenAI kreves.", "or": "eller", - "Organize your users": "", + "Organize your users": "Organisere brukerne dine", "Other": "Annet", "OUTPUT": "UTDATA", "Output format": "Format på utdata", "Overview": "Oversikt", "page": "side", "Password": "Passord", - "Paste Large Text as File": "", + "Paste Large Text as File": "Lim inn mye tekst som fil", "PDF document (.pdf)": "PDF-dokument (.pdf)", "PDF Extract Images (OCR)": "Uthenting av PDF-bilder (OCR)", "pending": "avventer", "Permission denied when accessing media devices": "Tilgang avslått ved bruk av medieenheter", "Permission denied when accessing microphone": "Tilgang avslått ved bruk av mikrofonen", "Permission denied when accessing microphone: {{error}}": "Tilgang avslått ved bruk av mikrofonen: {{error}}", - "Permissions": "", + "Permissions": "Tillatelser", "Personalization": "Tilpassing", "Pin": "Fest", "Pinned": "Festet", @@ -681,17 +679,17 @@ "Previous 30 days": "Siste 30 dager", "Previous 7 days": "Siste 7 dager", "Profile Image": "Profilbilde", - "Prompt (e.g. Tell me a fun fact about the Roman Empire)": "Ledetekst (f.eks. Fortell meg en morsom fakta om romerriket)", + "Prompt (e.g. Tell me a fun fact about the Roman Empire)": "Ledetekst (f.eks. Fortell meg noe morsomt om romerriket)", "Prompt Content": "Ledetekstinnhold", - "Prompt created successfully": "", + "Prompt created successfully": "Ledetekst opprettet", "Prompt suggestions": "Forslag til ledetekst", - "Prompt updated successfully": "", + "Prompt updated successfully": "Ledetekst oppdatert", "Prompts": "Ledetekster", - "Prompts Access": "", - "Proxy URL": "", - "Pull \"{{searchValue}}\" from Ollama.com": "Hent \"{{searchValue}}\" fra Ollama.com", + "Prompts Access": "Tilgang til ledetekster", + "Proxy URL": "Proxy-URL", + "Pull \"{{searchValue}}\" from Ollama.com": "Hent {{searchValue}} fra Ollama.com", "Pull a model from Ollama.com": "Hent en modell fra Ollama.com", - "Query Generation Prompt": "", + "Query Generation Prompt": "Ledetekst for genering av spørringer", "Query Params": "Spørringsparametere", "RAG Template": "RAG-mal", "Rating": "Vurdering", @@ -709,7 +707,7 @@ "Remove": "Fjern", "Remove Model": "Fjern modell", "Rename": "Gi nytt navn", - "Reorder Models": "", + "Reorder Models": "Sorter modeller på nytt", "Repeat Last N": "Gjenta siste N", "Request Mode": "Forespørselsmodus", "Reranking Model": "Omrangeringsmodell", @@ -718,7 +716,7 @@ "Reset": "Tilbakestill", "Reset All Models": "", "Reset Upload Directory": "Tilbakestill opplastingskatalog", - "Reset Vector Storage/Knowledge": "Tilbakestill Vector lagring/kunnskap", + "Reset Vector Storage/Knowledge": "Tilbakestill Vector-lagring/kunnskap", "Response notifications cannot be activated as the website permissions have been denied. Please visit your browser settings to grant the necessary access.": "Svar-varsler kan ikke aktiveres fordi tilgang til nettstedet er nektet. Gå til nettleserinnstillingene dine for å gi den nødvendige tilgangen.", "Response splitting": "Oppdeling av svar", "Result": "Resultat", @@ -755,7 +753,7 @@ "Search the web": "Søk på nettet", "Search Tools": "Søkeverktøy", "SearchApi API Key": "API-nøkkel for SearchApi", - "SearchApi Engine": "SearchApi-motor", + "SearchApi Engine": "Motor for SearchApi", "Searched {{count}} sites_one": "Søkte i {{count}} sites_one", "Searched {{count}} sites_other": "Søkte i {{count}} sites_other", "Searching \"{{searchQuery}}\"": "Søker etter \"{{searchQuery}}\"", @@ -767,7 +765,7 @@ "Select a base model": "Velg en grunnmodell", "Select a engine": "Velg en motor", "Select a function": "Velg en funksjon", - "Select a group": "", + "Select a group": "Velg en gruppe", "Select a model": "Velg en modell", "Select a pipeline": "Velg en pipeline", "Select a pipeline url": "Velg en pipeline-URL", @@ -790,7 +788,7 @@ "Set as default": "Angi som standard", "Set CFG Scale": "Angi CFG-skala", "Set Default Model": "Angi standard modell", - "Set embedding model": "", + "Set embedding model": "Angi innbyggingsmodell", "Set embedding model (e.g. {{model}})": "Angi innbyggingsmodell (f.eks. {{model}})", "Set Image Size": "Angi bildestørrelse", "Set reranking model (e.g. {{model}})": "Angi modell for omrangering (f.eks. {{model}})", @@ -828,7 +826,7 @@ "Source": "Kilde", "Speech Playback Speed": "Hastighet på avspilling av tale", "Speech recognition error: {{error}}": "Feil ved talegjenkjenning: {{error}}", - "Speech-to-Text Engine": "Tale-til-tekst-motor", + "Speech-to-Text Engine": "Motor for Tale-til-tekst", "Stop": "Stopp", "Stop Sequence": "Stoppsekvens", "Stream Chat Response": "Strømme chat-svar", @@ -857,7 +855,7 @@ "Text-to-Speech Engine": "Tekst-til-tale-motor", "Tfs Z": "Tfs Z", "Thanks for your feedback!": "Takk for tilbakemeldingen!", - "The Application Account DN you bind with for search": "Applikasjonskontoens DN du binder deg med for søk", + "The Application Account DN you bind with for search": "Applikasjonskontoens DN du binder deg med for søking", "The base to search for users": "Basen for å søke etter brukere", "The batch size determines how many text requests are processed together at once. A higher batch size can increase the performance and speed of the model, but it also requires more memory. (Default: 512)": "Batchstørrelsen avgjør hvor mange tekstforespørsler som behandles samtidig. En høyere batchstørrelse kan øke ytelsen og hastigheten til modellen, men det krever også mer minne. (Standard: 512)", "The developers behind this plugin are passionate volunteers from the community. If you find this plugin helpful, please consider contributing to its development.": "Utviklerne bak denne utvidelsen er lidenskapelige frivillige fra fellesskapet. Hvis du finner denne utvidelsen nyttig, vennligst vurder å bidra til utviklingen.", @@ -879,18 +877,18 @@ "This response was generated by \"{{model}}\"": "Dette svaret er generert av \"{{modell}}\"", "This will delete": "Dette sletter", "This will delete {{NAME}} and all its contents.": "Dette sletter {{NAME}} og alt innholdet.", - "This will delete all models including custom models": "", - "This will delete all models including custom models and cannot be undone.": "", + "This will delete all models including custom models": "Dette sletter alle modeller, inkludert tilpassede modeller", + "This will delete all models including custom models and cannot be undone.": "Dette sletter alle modeller, inkludert tilpassede modeller, og kan ikke angres.", "This will reset the knowledge base and sync all files. Do you wish to continue?": "Dette tilbakestiller kunnskapsbasen og synkroniserer alle filer. Vil du fortsette?", "Thorough explanation": "Grundig forklaring", "Tika": "Tika", - "Tika Server URL required.": "Tika server-URL kreves.", + "Tika Server URL required.": "Server-URL for Tika kreves.", "Tiktoken": "Tiktoken", "Tip: Update multiple variable slots consecutively by pressing the tab key in the chat input after each replacement.": "Tips: Oppdater flere variabelplasser etter hverandre ved å trykke på TAB-tasten i chat-inntastingsfeltet etter hver erstatning.", "Title": "Tittel", "Title (e.g. Tell me a fun fact)": "Tittel (f.eks. Fortell meg noe morsomt)", "Title Auto-Generation": "Automatisk tittelgenerering", - "Title cannot be an empty string.": "Tittelen kan ikke være en tom streng.", + "Title cannot be an empty string.": "Tittel kan ikke være en tom streng.", "Title Generation Prompt": "Ledetekst for tittelgenerering", "TLS": "TLS", "To access the available model names for downloading,": "Hvis du vil ha tilgang til modellnavn tilgjengelige for nedlasting,", @@ -910,13 +908,13 @@ "Too verbose": "For omfattende", "Tool created successfully": "Verktøy opprettet", "Tool deleted successfully": "Verktøy slettet", - "Tool Description": "", - "Tool ID": "", + "Tool Description": "Verktøyets beskrivelse", + "Tool ID": "Verktøyets ID", "Tool imported successfully": "Verktøy importert", - "Tool Name": "", + "Tool Name": "Verktøyets navn", "Tool updated successfully": "Verktøy oppdatert", "Tools": "Verktøy", - "Tools Access": "", + "Tools Access": "Verktøyets tilgang", "Tools are a function calling system with arbitrary code execution": "Verktøy er et funksjonskallsystem med vilkårlig kodekjøring", "Tools have a function calling system that allows arbitrary code execution": "Verktøy inneholder et funksjonskallsystem som tillater vilkårlig kodekjøring", "Tools have a function calling system that allows arbitrary code execution.": "Verktøy inneholder et funksjonskallsystem som tillater vilkårlig kodekjøring.", @@ -954,9 +952,9 @@ "Upload Progress": "Opplastingsfremdrift", "URL": "URL", "URL Mode": "URL-modus", - "Use '#' in the prompt input to load and include your knowledge.": "Bruk # i ledetekstinndata for å laste inn og inkludere dine kunnskaper.", + "Use '#' in the prompt input to load and include your knowledge.": "Bruk # i ledetekstens inntastingsfelt for å laste inn og inkludere kunnskapene dine.", "Use Gravatar": "Bruk Gravatar", - "Use groups to group your users and assign permissions.": "", + "Use groups to group your users and assign permissions.": "Bruk grupper til å samle brukere og tildele tillatelser.", "Use Initials": "Bruk initialer", "use_mlock (Ollama)": "use_mlock (Ollama)", "use_mmap (Ollama)": "use_mmap (Ollama)", @@ -975,12 +973,12 @@ "variable to have them replaced with clipboard content.": "variabel for å erstatte dem med utklippstavleinnhold.", "Version": "Versjon", "Version {{selectedVersion}} of {{totalVersions}}": "Version {{selectedVersion}} av {{totalVersions}}", - "Visibility": "", + "Visibility": "Synlighet", "Voice": "Stemme", "Voice Input": "Taleinndata", "Warning": "Advarsel", "Warning:": "Advarsel!", - "Warning: Enabling this will allow users to upload arbitrary code on the server.": "", + "Warning: Enabling this will allow users to upload arbitrary code on the server.": "Advarsel: Hvis du aktiverer denne funksjonen, kan brukere laste opp vilkårlig kode på serveren.", "Warning: If you update or change your embedding model, you will need to re-import all documents.": "Advarsel: Hvis du oppdaterer eller endrer innbyggingsmodellen din, må du importere alle dokumenter på nytt.", "Web": "Web", "Web API": "Web-API", @@ -998,13 +996,13 @@ "When enabled, the model will respond to each chat message in real-time, generating a response as soon as the user sends a message. This mode is useful for live chat applications, but may impact performance on slower hardware.": "Hvis denne modusen er aktivert, svarer modellen på alle chattemeldinger i sanntid, og genererer et svar så snart brukeren sender en melding. Denne modusen er nyttig for live chat-applikasjoner, men kan påvirke ytelsen på tregere maskinvare.", "wherever you are": "uansett hvor du er", "Whisper (Local)": "Whisper (Lokal)", - "Why?": "", + "Why?": "Hvorfor?", "Widescreen Mode": "Bredskjermmodus", "Won": "Vant", "Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. (Default: 0.9)": "Fungerer sammen med top-k. En høyere verdi (f.eks. 0,95) vil føre til mer mangfoldig tekst, mens en lavere verdi (f.eks. 0,5) vil generere mer fokusert og konservativ tekst. (Standard: 0,9)", "Workspace": "Arbeidsområde", - "Workspace Permissions": "", - "Write a prompt suggestion (e.g. Who are you?)": "Skriv inn et ledetekstforslag (f.eks. Hvem er du?)", + "Workspace Permissions": "Tillatelser for arbeidsområde", + "Write a prompt suggestion (e.g. Who are you?)": "Skriv inn et forslag til ledetekst (f.eks. Hvem er du?)", "Write a summary in 50 words that summarizes [topic or keyword].": "Skriv inn et sammendrag på 50 ord som oppsummerer [emne eller nøkkelord].", "Write something...": "Skriv inn noe...", "Write your model template content here": "Skriv inn modellens malinnhold her", @@ -1013,12 +1011,12 @@ "You can only chat with a maximum of {{maxCount}} file(s) at a time.": "Du kan bare chatte med maksimalt {{maxCount}} fil(er) om gangen.", "You can personalize your interactions with LLMs by adding memories through the 'Manage' button below, making them more helpful and tailored to you.": "Du kan tilpasse interaksjonene dine med språkmodeller ved å legge til minner gjennom Administrer-knappen nedenfor, slik at de blir mer til nyttige og tilpasset deg.", "You cannot upload an empty file.": "Du kan ikke laste opp en tom fil.", - "You do not have permission to upload files.": "", + "You do not have permission to upload files.": "Du har ikke tillatelse til å laste opp filer.", "You have no archived conversations.": "Du har ingen arkiverte samtaler.", "You have shared this chat": "Du har delt denne chatten", "You're a helpful assistant.": "Du er en nyttig assistent.", "You're now logged in.": "Du er nå logget inn.", - "Your account status is currently pending activation.": "Status på kontoen er for øyeblikket ventende på aktivering.", + "Your account status is currently pending activation.": "Status på kontoen din er for øyeblikket ventende på aktivering.", "Your entire contribution will go directly to the plugin developer; Open WebUI does not take any percentage. However, the chosen funding platform might have its own fees.": "Hele beløpet går uavkortet til utvikleren av tillegget. Open WebUI mottar ikke deler av beløpet. Den valgte betalingsplattformen kan ha gebyrer.", "Youtube": "Youtube", "Youtube Loader Settings": "Innstillinger for YouTube-laster" diff --git a/src/lib/i18n/locales/sk-SK/translation.json b/src/lib/i18n/locales/sk-SK/translation.json new file mode 100644 index 000000000..0e9e3bc78 --- /dev/null +++ b/src/lib/i18n/locales/sk-SK/translation.json @@ -0,0 +1,1027 @@ +{ + "-1 for no limit, or a positive integer for a specific limit": "", + "'s', 'm', 'h', 'd', 'w' or '-1' for no expiration.": "'s', 'm', 'h', 'd', 'w' alebo '-1' pre žiadne vypršanie platnosti.", + "(e.g. `sh webui.sh --api --api-auth username_password`)": "(napr. `sh webui.sh --api --api-auth username_password`)", + "(e.g. `sh webui.sh --api`)": "(napr. `sh webui.sh --api`)", + "(latest)": "Najnovšie", + "{{ models }}": "{{ models }}", + "{{user}}'s Chats": "{{user}}'s konverzácie", + "{{webUIName}} Backend Required": "Vyžaduje sa {{webUIName}} Backend", + "*Prompt node ID(s) are required for image generation": "*Sú potrebné IDs pre prompt node na generovanie obrázkov", + "A new version (v{{LATEST_VERSION}}) is now available.": "Nová verzia (v{{LATEST_VERSION}}) je teraz k dispozícii.", + "A task model is used when performing tasks such as generating titles for chats and web search queries": "Model úloh sa používa pri vykonávaní úloh, ako je generovanie názvov pre chaty a vyhľadávacie dotazy na webe.", + "a user": "užívateľ", + "About": "O programe", + "Access": "Prístup", + "Access Control": "", + "Accessible to all users": "Prístupné pre všetkých užívateľov", + "Account": "Účet", + "Account Activation Pending": "Čaká sa na aktiváciu účtu", + "Accurate information": "Presné informácie", + "Actions": "Akcie", + "Activate this command by typing \"/{{COMMAND}}\" to chat input.": "Aktivujte tento príkaz napísaním \"/{{COMMAND}}\" do chatového vstupu", + "Active Users": "Aktívni užívatelia", + "Add": "Pridať", + "Add a model ID": "Pridať ID modelu", + "Add a short description about what this model does": "Pridajte krátky popis toho, čo tento model robí.", + "Add a tag": "Pridať štítok", + "Add Arena Model": "Pridať Arena model", + "Add Connection": "Pridať pripojenie", + "Add Content": "Pridať obsah", + "Add content here": "Pridať obsah sem", + "Add custom prompt": "Pridanie vlastného promptu", + "Add Files": "Pridať súbory", + "Add Group": "Pridať skupinu", + "Add Memory": "Pridať pamäť", + "Add Model": "Pridať model", + "Add Tag": "Pridať štítok", + "Add Tags": "Pridať štítky", + "Add text content": "Pridajte textový obsah", + "Add User": "Pridať užívateľa", + "Add User Group": "Pridať skupinu užívateľov", + "Adjusting these settings will apply changes universally to all users.": "Úprava týchto nastavení sa prejaví univerzálne u všetkých užívateľov.", + "admin": "admin", + "Admin": "Admin", + "Admin Panel": "Admin panel", + "Admin Settings": "Nastavenia admina", + "Admins have access to all tools at all times; users need tools assigned per model in the workspace.": "Administrátori majú prístup ku všetkým nástrojom kedykoľvek; užívatelia potrebujú mať nástroje priradené podľa modelu v workspace.", + "Advanced Parameters": "Pokročilé parametre", + "Advanced Params": "Pokročilé parametre", + "All chats": "Všetky konverzácie", + "All Documents": "Všetky dokumenty", + "All models deleted successfully": "Všetky modely úspešne odstránené", + "Allow Chat Delete": "Povoliť odstránenie chatu", + "Allow Chat Deletion": "Povoliť odstránenie chatu", + "Allow Chat Edit": "Povoliť úpravu chatu", + "Allow File Upload": "Povoliť nahrávanie súborov", + "Allow non-local voices": "Povoliť ne-lokálne hlasy", + "Allow Temporary Chat": "Povoliť dočasný chat", + "Allow User Location": "Povoliť užívateľskú polohu", + "Allow Voice Interruption in Call": "Povoliť prerušenie hlasu počas hovoru", + "Already have an account?": "Už máte účet?", + "Alternative to the top_p, and aims to ensure a balance of quality and variety. The parameter p represents the minimum probability for a token to be considered, relative to the probability of the most likely token. For example, with p=0.05 and the most likely token having a probability of 0.9, logits with a value less than 0.045 are filtered out. (Default: 0.0)": "", + "Amazing": "", + "an assistant": "asistent", + "and": "a", + "and {{COUNT}} more": "a {{COUNT}} ďalšie/í", + "and create a new shared link.": "a vytvoriť nový zdieľaný odkaz.", + "API Base URL": "Základná URL adresa API", + "API Key": "API kľúč", + "API Key created.": "API kľúč bol vytvorený.", + "API keys": "API kľúče", + "Application DN": "", + "Application DN Password": "", + "applies to all users with the \"user\" role": "", + "April": "Apríl", + "Archive": "Archivovať", + "Archive All Chats": "Archivovať všetky konverzácie", + "Archived Chats": "Archivované konverzácie", + "archived-chat-export": "", + "Are you sure you want to unarchive all archived chats?": "", + "Are you sure?": "Ste si istý?", + "Arena Models": "Arena modely", + "Artifacts": "Artefakty", + "Ask a question": "Opýtajte sa otázku", + "Assistant": "Asistent", + "Attach file": "Pripojiť súbor", + "Attention to detail": "Pozornosť k detailom", + "Attribute for Username": "", + "Audio": "Zvuk", + "August": "August", + "Authenticate": "Autentifikovať", + "Auto-Copy Response to Clipboard": "Automatické kopírovanie odpovede do schránky", + "Auto-playback response": "Automatická odpoveď pri prehrávaní", + "Autocomplete Generation": "", + "Autocomplete Generation Input Max Length": "", + "Automatic1111": "Automatic1111", + "AUTOMATIC1111 Api Auth String": "AUTOMATIC1111 Api Auth String", + "AUTOMATIC1111 Base URL": "Základná URL pre AUTOMATIC1111", + "AUTOMATIC1111 Base URL is required.": "Vyžaduje sa základná URL pre AUTOMATIC1111.", + "Available list": "Dostupný zoznam", + "available!": "k dispozícii!", + "Awful": "", + "Azure AI Speech": "Azure AI syntéza reči", + "Azure Region": "Azure oblasť", + "Back": "Späť", + "Bad Response": "Zlá odozva", + "Banners": "Bannery", + "Base Model (From)": "Základný model (z)", + "Batch Size (num_batch)": "Veľkosť batchu (num_batch)", + "before": "pred", + "Being lazy": "", + "Bing Search V7 Endpoint": "", + "Bing Search V7 Subscription Key": "", + "Brave Search API Key": "API kľúč pre Brave Search", + "By {{name}}": "", + "Bypass SSL verification for Websites": "Obísť overenie SSL pre webové stránky", + "Call": "Volanie", + "Call feature is not supported when using Web STT engine": "Funkcia volania nie je podporovaná pri použití Web STT engine.", + "Camera": "Kamera", + "Cancel": "Zrušiť", + "Capabilities": "Schopnosti", + "Certificate Path": "", + "Change Password": "Zmeniť heslo", + "Character": "Znak", + "Character limit for autocomplete generation input": "", + "Chart new frontiers": "", + "Chat": "Chat", + "Chat Background Image": "Obrázok pozadia chatu", + "Chat Bubble UI": "Používateľské rozhranie bublín chatu (Chat Bubble UI)", + "Chat Controls": "Ovládanie chatu", + "Chat direction": "Smer chatu", + "Chat Overview": "Prehľad chatu", + "Chat Permissions": "", + "Chat Tags Auto-Generation": "Automatické generovanie značiek chatu", + "Chats": "Chaty", + "Check Again": "Skontroluj znovu", + "Check for updates": "Skontrolovať aktualizácie", + "Checking for updates...": "Kontrola aktualizácií...", + "Choose a model before saving...": "Vyberte model pred uložením...", + "Chunk Overlap": "", + "Chunk Params": "", + "Chunk Size": "", + "Ciphers": "", + "Citation": "Odkaz", + "Clear memory": "Vymazať pamäť", + "click here": "", + "Click here for filter guides.": "", + "Click here for help.": "Kliknite tu pre pomoc.", + "Click here to": "Kliknite tu na", + "Click here to download user import template file.": "Kliknite tu pre stiahnutie šablóny súboru na import užívateľov.", + "Click here to learn more about faster-whisper and see the available models.": "Kliknite sem a dozviete sa viac o faster-whisper a pozrite si dostupné modely.", + "Click here to select": "Kliknite sem pre výber", + "Click here to select a csv file.": "Kliknite sem pre výber súboru typu csv.", + "Click here to select a py file.": "Kliknite sem pre výber {{py}} súboru.", + "Click here to upload a workflow.json file.": "Kliknite sem pre nahratie súboru workflow.json.", + "click here.": "kliknite tu.", + "Click on the user role button to change a user's role.": "Kliknite na tlačidlo role užívateľa, aby ste zmenili rolu užívateľa.", + "Clipboard write permission denied. Please check your browser settings to grant the necessary access.": "Prístup na zápis do schránky bol zamietnutý. Skontrolujte nastavenia prehliadača a udeľte potrebný prístup.", + "Clone": "Klonovať", + "Close": "Zavrieť", + "Code execution": "Vykonávanie kódu", + "Code formatted successfully": "Kód bol úspešne naformátovaný.", + "Collection": "", + "Color": "Farba", + "ComfyUI": "ComfyUI", + "ComfyUI Base URL": "Základná URL ComfyUI", + "ComfyUI Base URL is required.": "Je vyžadovaná základná URL pre ComfyUI.", + "ComfyUI Workflow": "Pracovný postup ComfyUI", + "ComfyUI Workflow Nodes": "Pracovné uzly ComfyUI", + "Command": "Príkaz", + "Completions": "Doplnenia", + "Concurrent Requests": "Súčasné požiadavky", + "Configure": "Konfigurovať", + "Configure Models": "Konfigurovať modely", + "Confirm": "Potvrdiť", + "Confirm Password": "Potvrdenie hesla", + "Confirm your action": "Potvrďte svoju akciu", + "Connections": "Pripojenia", + "Contact Admin for WebUI Access": "Kontaktujte administrátora pre prístup k webovému rozhraniu.", + "Content": "Obsah", + "Content Extraction": "Extrakcia obsahu", + "Context Length": "Dĺžka kontextu", + "Continue Response": "Pokračovať v odpovedi", + "Continue with {{provider}}": "Pokračovať s {{provider}}", + "Continue with Email": "", + "Continue with LDAP": "", + "Control how message text is split for TTS requests. 'Punctuation' splits into sentences, 'paragraphs' splits into paragraphs, and 'none' keeps the message as a single string.": "Kontrola, ako sa text správy rozdeľuje pre požiadavky TTS. 'Punctuation' rozdeľuje text na vety, 'paragraphs' rozdeľuje text na odseky a 'none' ponecháva správu ako jeden celý reťazec.", + "Controls": "Ovládacie prvky", + "Controls the balance between coherence and diversity of the output. A lower value will result in more focused and coherent text. (Default: 5.0)": "", + "Copied": "Skopírované", + "Copied shared chat URL to clipboard!": "URL zdieľanej konverzácie skopírované do schránky!", + "Copied to clipboard": "Skopírované do schránky", + "Copy": "Kopírovať", + "Copy last code block": "Skopírujte posledný blok kódu", + "Copy last response": "Skopírujte poslednú odpoveď", + "Copy Link": "Kopírovať odkaz", + "Copy to clipboard": "Kopírovať do schránky", + "Copying to clipboard was successful!": "Kopírovanie do schránky bolo úspešné!", + "Create": "Vytvoriť", + "Create a knowledge base": "Vytvoriť knowledge base", + "Create a model": "Vytvoriť model", + "Create Account": "Vytvoriť účet", + "Create Admin Account": "Vytvoriť admin účet", + "Create Group": "Vytvoriť skupinu", + "Create Knowledge": "Vytvoriť knowledge", + "Create new key": "Vytvoriť nový kľúč", + "Create new secret key": "Vytvoriť nový tajný kľúč", + "Created at": "Vytvorené dňa", + "Created At": "Vytvorené dňa", + "Created by": "Vytvorené užívateľom", + "CSV Import": "CSV import", + "Current Model": "Aktuálny model", + "Current Password": "Aktuálne heslo", + "Custom": "Na mieru", + "Dark": "Tmavý", + "Database": "Databáza", + "December": "December", + "Default": "Predvolené hodnoty alebo nastavenia.", + "Default (Open AI)": "Predvolené (Open AI)", + "Default (SentenceTransformers)": "Predvolené (SentenceTransformers)", + "Default Model": "Predvolený model", + "Default model updated": "Predvolený model aktualizovaný.", + "Default Models": "Predvolené modely", + "Default permissions": "Predvolené povolenia", + "Default permissions updated successfully": "Predvolené povolenia úspešne aktualizované", + "Default Prompt Suggestions": "Predvolené návrhy promptov", + "Default to 389 or 636 if TLS is enabled": "", + "Default to ALL": "", + "Default User Role": "Predvolená rola užívateľa", + "Delete": "Odstrániť", + "Delete a model": "Odstrániť model.", + "Delete All Chats": "Odstrániť všetky konverzácie", + "Delete All Models": "", + "Delete chat": "Odstrániť chat", + "Delete Chat": "Odstrániť chat", + "Delete chat?": "Odstrániť konverzáciu?", + "Delete folder?": "Odstrániť priečinok?", + "Delete function?": "Funkcia na odstránenie?", + "Delete prompt?": "Odstrániť prompt?", + "delete this link": "odstrániť tento odkaz", + "Delete tool?": "Odstrániť nástroj?", + "Delete User": "Odstrániť užívateľa", + "Deleted {{deleteModelTag}}": "Odstránené {{deleteModelTag}}", + "Deleted {{name}}": "Odstránené {{name}}", + "Deleted User": "", + "Describe your knowledge base and objectives": "", + "Description": "Popis", + "Didn't fully follow instructions": "Nenasledovali ste presne všetky inštrukcie.", + "Disabled": "Zakázané", + "Discover a function": "Objaviť funkciu", + "Discover a model": "Objaviť model", + "Discover a prompt": "Objaviť prompt", + "Discover a tool": "Objaviť nástroj", + "Discover wonders": "", + "Discover, download, and explore custom functions": "Objavujte, sťahujte a preskúmajte vlastné funkcie", + "Discover, download, and explore custom prompts": "Objavte, stiahnite a preskúmajte vlastné prompty.", + "Discover, download, and explore custom tools": "Objavujte, sťahujte a preskúmajte vlastné nástroje", + "Discover, download, and explore model presets": "Objavte, stiahnite a preskúmajte prednastavenia modelov", + "Dismissible": "Odstrániteľné", + "Display": "", + "Display Emoji in Call": "Zobrazenie emoji počas hovoru", + "Display the username instead of You in the Chat": "Zobraziť užívateľské meno namiesto \"Vás\" v chate", + "Displays citations in the response": "", + "Dive into knowledge": "", + "Do not install functions from sources you do not fully trust.": "Neinštalujte funkcie zo zdrojov, ktorým plne nedôverujete.", + "Do not install tools from sources you do not fully trust.": "Neinštalujte nástroje zo zdrojov, ktorým plne nedôverujete.", + "Document": "Dokument", + "Documentation": "Dokumentácia", + "Documents": "Dokumenty", + "does not make any external connections, and your data stays securely on your locally hosted server.": "nevytvára žiadne externé pripojenia a vaše dáta zostávajú bezpečne na vašom lokálnom serveri.", + "Don't have an account?": "Nemáte účet?", + "don't install random functions from sources you don't trust.": "Neinštalujte náhodné funkcie zo zdrojov, ktorým nedôverujete.", + "don't install random tools from sources you don't trust.": "Neinštalujte náhodné nástroje zo zdrojov, ktorým nedôverujete.", + "Don't like the style": "Nepáči sa mi tento štýl.", + "Done": "Hotovo.", + "Download": "Stiahnuť", + "Download canceled": "Sťahovanie zrušené", + "Download Database": "Stiahnuť databázu", + "Drag and drop a file to upload or select a file to view": "", + "Draw": "Nakresliť", + "Drop any files here to add to the conversation": "Sem presuňte akékoľvek súbory, ktoré chcete pridať do konverzácie", + "e.g. '30s','10m'. Valid time units are 's', 'm', 'h'.": "napr. '30s','10m'. Platné časové jednotky sú 's', 'm', 'h'.", + "e.g. A filter to remove profanity from text": "", + "e.g. My Filter": "", + "e.g. My Tools": "", + "e.g. my_filter": "", + "e.g. my_tools": "", + "e.g. Tools for performing various operations": "", + "Edit": "Upraviť", + "Edit Arena Model": "Upraviť Arena Model", + "Edit Connection": "", + "Edit Default Permissions": "", + "Edit Memory": "Upraviť pamäť", + "Edit User": "Upraviť užívateľa", + "Edit User Group": "", + "ElevenLabs": "ElevenLabs", + "Email": "E-mail", + "Embark on adventures": "", + "Embedding Batch Size": "", + "Embedding Model": "Vkladací model (Embedding Model)", + "Embedding Model Engine": "", + "Embedding model set to \"{{embedding_model}}\"": "Model vkladania nastavený na \"{{embedding_model}}\"", + "Enable API Key Auth": "", + "Enable autocomplete generation for chat messages": "", + "Enable Community Sharing": "Povoliť zdieľanie komunity", + "Enable Memory Locking (mlock) to prevent model data from being swapped out of RAM. This option locks the model's working set of pages into RAM, ensuring that they will not be swapped out to disk. This can help maintain performance by avoiding page faults and ensuring fast data access.": "", + "Enable Memory Mapping (mmap) to load model data. This option allows the system to use disk storage as an extension of RAM by treating disk files as if they were in RAM. This can improve model performance by allowing for faster data access. However, it may not work correctly with all systems and can consume a significant amount of disk space.": "", + "Enable Message Rating": "Povoliť hodnotenie správ", + "Enable Mirostat sampling for controlling perplexity. (Default: 0, 0 = Disabled, 1 = Mirostat, 2 = Mirostat 2.0)": "", + "Enable New Sign Ups": "Povoliť nové registrácie", + "Enable Web Search": "Povoliť webové vyhľadávanie", + "Enabled": "Povolené", + "Engine": "Engine", + "Ensure your CSV file includes 4 columns in this order: Name, Email, Password, Role.": "Uistite sa, že váš CSV súbor obsahuje 4 stĺpce v tomto poradí: Name, Email, Password, Role.", + "Enter {{role}} message here": "Zadajte správu {{role}} sem", + "Enter a detail about yourself for your LLMs to recall": "Zadajte podrobnosť o sebe, ktorú si vaše LLM majú zapamätať.", + "Enter api auth string (e.g. username:password)": "Zadajte autentifikačný reťazec API (napr. užívateľské_meno:heslo)", + "Enter Application DN": "", + "Enter Application DN Password": "", + "Enter Bing Search V7 Endpoint": "", + "Enter Bing Search V7 Subscription Key": "", + "Enter Brave Search API Key": "Zadajte API kľúč pre Brave Search", + "Enter certificate path": "", + "Enter CFG Scale (e.g. 7.0)": "Zadajte mierku CFG (napr. 7.0)", + "Enter Chunk Overlap": "Zadajte prekryv časti", + "Enter Chunk Size": "Zadajte veľkosť časti", + "Enter description": "Zadajte popis", + "Enter Github Raw URL": "Zadajte URL adresu Github Raw", + "Enter Google PSE API Key": "Zadajte kľúč rozhrania API Google PSE", + "Enter Google PSE Engine Id": "Zadajte ID vyhľadávacieho mechanizmu Google PSE", + "Enter Image Size (e.g. 512x512)": "Zadajte veľkosť obrázka (napr. 512x512)", + "Enter Jina API Key": "", + "Enter language codes": "Zadajte kódy jazykov", + "Enter Model ID": "Zadajte ID modelu", + "Enter model tag (e.g. {{modelTag}})": "Zadajte označenie modelu (napr. {{modelTag}})", + "Enter Mojeek Search API Key": "", + "Enter Number of Steps (e.g. 50)": "Zadajte počet krokov (napr. 50)", + "Enter proxy URL (e.g. https://user:password@host:port)": "", + "Enter Sampler (e.g. Euler a)": "Zadajte vzorkovač (napr. Euler a)", + "Enter Scheduler (e.g. Karras)": "Zadajte plánovač (napr. Karras)", + "Enter Score": "Zadajte skóre", + "Enter SearchApi API Key": "Zadajte API kľúč pre SearchApi", + "Enter SearchApi Engine": "Zadajte vyhľadávací engine SearchApi", + "Enter Searxng Query URL": "Zadajte URL dopytu Searxng", + "Enter Seed": "", + "Enter Serper API Key": "Zadajte Serper API kľúč", + "Enter Serply API Key": "Zadajte API kľúč pre Serply", + "Enter Serpstack API Key": "Zadajte kľúč API pre Serpstack", + "Enter server host": "", + "Enter server label": "", + "Enter server port": "", + "Enter stop sequence": "Zadajte ukončovaciu sekvenciu", + "Enter system prompt": "Vložte systémový prompt", + "Enter Tavily API Key": "Zadajte API kľúč Tavily", + "Enter Tika Server URL": "Zadajte URL servera Tika", + "Enter Top K": "Zadajte horné K", + "Enter URL (e.g. http://127.0.0.1:7860/)": "Zadajte URL (napr. http://127.0.0.1:7860/)", + "Enter URL (e.g. http://localhost:11434)": "Zadajte URL (napr. http://localhost:11434)", + "Enter Your Email": "Zadajte svoj email", + "Enter Your Full Name": "Zadajte svoje celé meno", + "Enter your message": "Zadajte svoju správu", + "Enter Your Password": "Zadajte svoje heslo", + "Enter Your Role": "Zadajte svoju rolu", + "Enter Your Username": "", + "Error": "Chyba", + "ERROR": "Chyba", + "Evaluations": "Hodnotenia", + "Example: (&(objectClass=inetOrgPerson)(uid=%s))": "", + "Example: ALL": "", + "Example: ou=users,dc=foo,dc=example": "", + "Example: sAMAccountName or uid or userPrincipalName": "", + "Exclude": "Vylúčiť", + "Experimental": "Experimentálne", + "Explore the cosmos": "", + "Export": "Exportovať", + "Export All Archived Chats": "", + "Export All Chats (All Users)": "Exportovať všetky konverzácie (všetci užívatelia)", + "Export chat (.json)": "Exportovať konverzáciu (.json)", + "Export Chats": "Exportovať konverzácie", + "Export Config to JSON File": "Exportujte konfiguráciu do súboru JSON", + "Export Functions": "Exportovať funkcie", + "Export Models": "Exportovať modely", + "Export Presets": "", + "Export Prompts": "Exportovať prompty", + "Export to CSV": "", + "Export Tools": "Exportné nástroje", + "External Models": "Externé modely", + "Failed to add file.": "Nepodarilo sa pridať súbor.", + "Failed to create API Key.": "Nepodarilo sa vytvoriť API kľúč.", + "Failed to read clipboard contents": "Nepodarilo sa prečítať obsah schránky", + "Failed to save models configuration": "", + "Failed to update settings": "Nepodarilo sa aktualizovať nastavenia", + "Failed to upload file.": "Nepodarilo sa nahrať súbor.", + "February": "Február", + "Feedback History": "História spätnej väzby", + "Feedbacks": "", + "Feel free to add specific details": "Neváhajte pridať konkrétne detaily.", + "File": "Súbor", + "File added successfully.": "Súbor bol úspešne pridaný.", + "File content updated successfully.": "Obsah súboru bol úspešne aktualizovaný.", + "File Mode": "Režim súboru", + "File not found.": "Súbor nenájdený.", + "File removed successfully.": "Súbor bol úspešne odstránený.", + "File size should not exceed {{maxSize}} MB.": "Veľkosť súboru by nemala presiahnuť {{maxSize}} MB.", + "Files": "Súbory", + "Filter is now globally disabled": "Filter je teraz globálne zakázaný", + "Filter is now globally enabled": "Filter je teraz globálne povolený.", + "Filters": "Filtre", + "Fingerprint spoofing detected: Unable to use initials as avatar. Defaulting to default profile image.": "Zistené falšovanie odtlačkov prstov: Nie je možné použiť iniciály ako avatar. Používa sa predvolený profilový obrázok.", + "Fluidly stream large external response chunks": "Plynule streamujte veľké externé časti odpovedí", + "Focus chat input": "Zamerajte sa na vstup chatu", + "Folder deleted successfully": "Priečinok bol úspešne vymazaný", + "Folder name cannot be empty": "Názov priečinka nesmie byť prázdny", + "Folder name cannot be empty.": "Názov priečinka nesmie byť prázdny.", + "Folder name updated successfully": "Názov priečinka bol úspešne aktualizovaný.", + "Followed instructions perfectly": "Dodržal pokyny dokonale.", + "Forge new paths": "", + "Form": "Formulár", + "Format your variables using brackets like this:": "Formátujte svoje premenné pomocou zátvoriek takto:", + "Frequency Penalty": "Penalizácia frekvencie", + "Function": "Funkcia", + "Function created successfully": "Funkcia bola úspešne vytvorená.", + "Function deleted successfully": "Funkcia bola úspešne odstránená", + "Function Description": "", + "Function ID": "", + "Function is now globally disabled": "Funkcia je teraz globálne zakázaná.", + "Function is now globally enabled": "Funkcia je teraz globálne povolená.", + "Function Name": "", + "Function updated successfully": "Funkcia bola úspešne aktualizovaná.", + "Functions": "Funkcie", + "Functions allow arbitrary code execution": "Funkcie umožňujú vykonávať ľubovoľný kód.", + "Functions allow arbitrary code execution.": "Funkcie umožňujú vykonávanie ľubovoľného kódu.", + "Functions imported successfully": "Funkcie boli úspešne importované", + "General": "Všeobecné", + "General Settings": "Všeobecné nastavenia", + "Generate Image": "Vygenerovať obrázok", + "Generating search query": "Generovanie vyhľadávacieho dotazu", + "Generation Info": "Informácie o generácii", + "Get started": "", + "Get started with {{WEBUI_NAME}}": "", + "Global": "Globálne", + "Good Response": "Dobrá odozva", + "Google PSE API Key": "Kľúč API pre Google PSE (Programmatically Search Engine)", + "Google PSE Engine Id": "Google PSE Engine Id (Identifikátor vyhľadávacieho modulu Google PSE)", + "Group created successfully": "", + "Group deleted successfully": "", + "Group Description": "", + "Group Name": "", + "Group updated successfully": "", + "Groups": "", + "h:mm a": "hh:mm dop./odp.", + "Haptic Feedback": "Haptická spätná väzba", + "has no conversations.": "nemá žiadne konverzácie.", + "Hello, {{name}}": "Ahoj, {{name}}", + "Help": "Pomoc", + "Help us create the best community leaderboard by sharing your feedback history!": "Pomôžte nám vytvoriť najlepší komunitný rebríček zdieľaním histórie vašej spätnej väzby!", + "Hex Color": "", + "Hex Color - Leave empty for default color": "", + "Hide": "Skryť", + "Host": "", + "How can I help you today?": "Ako vám môžem dnes pomôcť?", + "How would you rate this response?": "", + "Hybrid Search": "Hybridné vyhľadávanie", + "I acknowledge that I have read and I understand the implications of my action. I am aware of the risks associated with executing arbitrary code and I have verified the trustworthiness of the source.": "Beriem na vedomie, že som si prečítal a chápem dôsledky svojich činov. Som si vedomý rizík spojených s vykonávaním ľubovoľného kódu a overil som dôveryhodnosť zdroja.", + "ID": "ID", + "Ignite curiosity": "", + "Image Generation (Experimental)": "Generovanie obrázkov (experimentálne)", + "Image Generation Engine": "Engine na generovanie obrázkov", + "Image Settings": "Nastavenia obrázka", + "Images": "Obrázky", + "Import Chats": "Importovať konverzácie", + "Import Config from JSON File": "Importovanie konfigurácie z JSON súboru", + "Import Functions": "Načítanie funkcií", + "Import Models": "Importovanie modelov", + "Import Presets": "", + "Import Prompts": "Importovať Prompty", + "Import Tools": "Importovať nástroje", + "Include": "Zahrnúť", + "Include `--api-auth` flag when running stable-diffusion-webui": "Zahrňte prepínač `--api-auth` pri spustení stable-diffusion-webui.", + "Include `--api` flag when running stable-diffusion-webui": "Pri spustení stable-diffusion-webui zahrňte príznak `--api`.", + "Influences how quickly the algorithm responds to feedback from the generated text. A lower learning rate will result in slower adjustments, while a higher learning rate will make the algorithm more responsive. (Default: 0.1)": "", + "Info": "Info", + "Input commands": "Vstupné príkazy", + "Install from Github URL": "Inštalácia z URL adresy Githubu", + "Instant Auto-Send After Voice Transcription": "Okamžité automatické odoslanie po prepisu hlasu", + "Interface": "Rozhranie", + "Invalid file format.": "Neplatný formát súboru.", + "Invalid Tag": "Neplatný tag", + "January": "Január", + "Jina API Key": "", + "join our Discord for help.": "pripojte sa k nášmu Discordu pre pomoc.", + "JSON": "JSON", + "JSON Preview": "Náhľad JSON", + "July": "Júl", + "June": "Jún", + "JWT Expiration": "Vypršanie platnosti JWT (JSON Web Token)", + "JWT Token": "JWT Token (JSON Web Token)", + "Keep Alive": "Udržiavať spojenie", + "Key": "", + "Keyboard shortcuts": "Klávesové skratky", + "Knowledge": "Znalosti", + "Knowledge Access": "", + "Knowledge created successfully.": "Znalosť úspešne vytvorená.", + "Knowledge deleted successfully.": "Znalosti boli úspešne odstránené.", + "Knowledge reset successfully.": "Úspešné obnovenie znalostí.", + "Knowledge updated successfully": "Znalosti úspešne aktualizované", + "Label": "", + "Landing Page Mode": "Režim vstupnej stránky", + "Language": "Jazyk", + "Last Active": "Naposledy aktívny", + "Last Modified": "Posledná zmena", + "LDAP": "", + "LDAP server updated": "", + "Leaderboard": "Rebríček", + "Leave empty for unlimited": "Nechajte prázdne pre neobmedzene", + "Leave empty to include all models from \"{{URL}}/api/tags\" endpoint": "", + "Leave empty to include all models from \"{{URL}}/models\" endpoint": "", + "Leave empty to include all models or select specific models": "Nechajte prázdne pre zahrnutie všetkých modelov alebo vyberte konkrétne modely.", + "Leave empty to use the default prompt, or enter a custom prompt": "Nechajte prázdne pre použitie predvoleného podnetu, alebo zadajte vlastný podnet.", + "Light": "Svetlo", + "Listening...": "Počúvanie...", + "LLMs can make mistakes. Verify important information.": "LLM môžu robiť chyby. Overte si dôležité informácie.", + "Local": "", + "Local Models": "Lokálne modely", + "Lost": "Stratený", + "LTR": "LTR", + "Made by OpenWebUI Community": "Vytvorené komunitou OpenWebUI", + "Make sure to enclose them with": "Uistite sa, že sú uzavreté pomocou", + "Make sure to export a workflow.json file as API format from ComfyUI.": "Uistite sa, že exportujete súbor workflow.json vo formáte API z ComfyUI.", + "Manage": "Spravovať", + "Manage Arena Models": "Správa modelov v Arena", + "Manage Ollama": "", + "Manage Ollama API Connections": "", + "Manage OpenAI API Connections": "", + "Manage Pipelines": "Správa pipelines", + "March": "Marec", + "Max Tokens (num_predict)": "Maximálny počet tokenov (num_predict)", + "Max Upload Count": "Maximálny počet nahraní", + "Max Upload Size": "Maximálna veľkosť nahrávania", + "Maximum of 3 models can be downloaded simultaneously. Please try again later.": "Maximálne 3 modely môžu byť stiahnuté súčasne. Prosím skúste to znova neskôr.", + "May": "Máj", + "Memories accessible by LLMs will be shown here.": "Spomienky prístupné LLM budú zobrazené tu.", + "Memory": "Pamäť", + "Memory added successfully": "Pamäť bola úspešne pridaná.", + "Memory cleared successfully": "Pamäť bola úspešne vymazaná.", + "Memory deleted successfully": "Pamäť bola úspešne vymazaná", + "Memory updated successfully": "Pamäť úspešne aktualizovaná", + "Merge Responses": "Zlúčiť odpovede", + "Message rating should be enabled to use this feature": "Hodnotenie správ musí byť povolené, aby bolo možné túto funkciu používať.", + "Messages you send after creating your link won't be shared. Users with the URL will be able to view the shared chat.": "Správy, ktoré odošlete po vytvorení odkazu, nebudú zdieľané. Používatelia s URL budú môcť zobraziť zdieľaný chat.", + "Min P": "Min P", + "Minimum Score": "Minimálne skóre", + "Mirostat": "Mirostat", + "Mirostat Eta": "Mirostat Eta", + "Mirostat Tau": "Mirostat Tau", + "MMMM DD, YYYY": "MMMM DD, RRRR", + "MMMM DD, YYYY HH:mm": "MMMM DD, RRRR HH:mm", + "MMMM DD, YYYY hh:mm:ss A": "MMMM DD, YYYY hh:mm:ss A", + "Model": "Model", + "Model '{{modelName}}' has been successfully downloaded.": "Model „{{modelName}}“ bol úspešne stiahnutý.", + "Model '{{modelTag}}' is already in queue for downloading.": "Model '{{modelTag}}' je už zaradený do fronty na sťahovanie.", + "Model {{modelId}} not found": "Model {{modelId}} nebol nájdený", + "Model {{modelName}} is not vision capable": "Model {{modelName}} nie je schopný spracovávať vizuálne údaje.", + "Model {{name}} is now {{status}}": "Model {{name}} je teraz {{status}}.", + "Model accepts image inputs": "Model prijíma vstupy vo forme obrázkov", + "Model created successfully!": "Model bol úspešne vytvorený!", + "Model filesystem path detected. Model shortname is required for update, cannot continue.": "Zistená cesta v súborovom systéme. Je vyžadovaný krátky názov modelu pre aktualizáciu, nemožno pokračovať.", + "Model Filtering": "", + "Model ID": "ID modelu", + "Model IDs": "", + "Model Name": "Názov modelu", + "Model not selected": "Model nebol vybraný", + "Model Params": "Parametre modelu", + "Model Permissions": "", + "Model updated successfully": "Model bol úspešne aktualizovaný", + "Modelfile Content": "Obsah súboru modelfile", + "Models": "Modely", + "Models Access": "", + "Models configuration saved successfully": "", + "Mojeek Search API Key": "", + "more": "viac", + "More": "Viac", + "Name": "Meno", + "Name your knowledge base": "", + "New Chat": "Nový chat", + "New folder": "Nový priečinok", + "New Password": "Nové heslo", + "No content found": "Nebol nájdený žiadny obsah.", + "No content to speak": "Žiadny obsah na diskusiu.", + "No distance available": "Nie je dostupná žiadna vzdialenosť", + "No feedbacks found": "Žiadna spätná väzba nenájdená", + "No file selected": "Nebola vybratá žiadna súbor", + "No files found.": "Neboli nájdené žiadne súbory.", + "No groups with access, add a group to grant access": "", + "No HTML, CSS, or JavaScript content found.": "Nebola nájdená žiadny obsah HTML, CSS ani JavaScript.", + "No knowledge found": "Neboli nájdené žiadne znalosti", + "No model IDs": "", + "No models found": "Neboli nájdené žiadne modely", + "No models selected": "", + "No results found": "Neboli nájdené žiadne výsledky", + "No search query generated": "Nebola vygenerovaná žiadna vyhľadávacia otázka.", + "No source available": "Nie je dostupný žiadny zdroj.", + "No users were found.": "", + "No valves to update": "Žiadne ventily na aktualizáciu", + "None": "Žiadny", + "Not factually correct": "Nie je fakticky správne", + "Not helpful": "Nepomocné", + "Note: If you set a minimum score, the search will only return documents with a score greater than or equal to the minimum score.": "Poznámka: Ak nastavíte minimálne skóre, vyhľadávanie vráti iba dokumenty s hodnotením, ktoré je väčšie alebo rovné zadanému minimálnemu skóre.", + "Notes": "Poznámky", + "Notifications": "Oznámenia", + "November": "November", + "num_gpu (Ollama)": "Počet GPU (Ollama)", + "num_thread (Ollama)": "num_thread (Ollama)", + "OAuth ID": "OAuth ID", + "October": "Október", + "Off": "Vypnuté", + "Okay, Let's Go!": "Dobre, poďme na to!", + "OLED Dark": "OLED Dark", + "Ollama": "Ollama", + "Ollama API": "Ollama API", + "Ollama API disabled": "API rozhranie Ollama je zakázané.", + "Ollama API settings updated": "", + "Ollama Version": "Verzia Ollama", + "On": "Zapnuté", + "Only alphanumeric characters and hyphens are allowed": "", + "Only alphanumeric characters and hyphens are allowed in the command string.": "Príkazový reťazec môže obsahovať iba alfanumerické znaky a pomlčky.", + "Only collections can be edited, create a new knowledge base to edit/add documents.": "Iba kolekcie môžu byť upravované, na úpravu/pridanie dokumentov vytvorte novú znalostnú databázu.", + "Only select users and groups with permission can access": "", + "Oops! Looks like the URL is invalid. Please double-check and try again.": "Jejda! Vyzerá to, že URL adresa je neplatná. Prosím, skontrolujte ju a skúste to znova.", + "Oops! There are files still uploading. Please wait for the upload to complete.": "Jejda! Niektoré súbory sa stále nahrávajú. Prosím, počkajte, kým sa nahrávanie dokončí.", + "Oops! There was an error in the previous response.": "Jejda! V predchádzajúcej odpovedi došlo k chybe.", + "Oops! You're using an unsupported method (frontend only). Please serve the WebUI from the backend.": "Jejda! Používate nepodporovanú metódu (iba frontend). Prosím, spustite WebUI zo serverovej časti (backendu).", + "Open file": "Otvoriť súbor", + "Open in full screen": "Otvoriť na celú obrazovku", + "Open new chat": "Otvoriť nový chat", + "Open WebUI uses faster-whisper internally.": "Open WebUI interne používa faster-whisper.", + "Open WebUI uses SpeechT5 and CMU Arctic speaker embeddings.": "", + "Open WebUI version (v{{OPEN_WEBUI_VERSION}}) is lower than required version (v{{REQUIRED_VERSION}})": "Verzia Open WebUI (v{{OPEN_WEBUI_VERSION}}) je nižšia ako požadovaná verzia (v{{REQUIRED_VERSION}})", + "OpenAI": "OpenAI je výskumná organizácia zameraná na umelú inteligenciu, ktorá je známa vývojom pokročilých jazykových modelov, ako je napríklad GPT. Tieto modely sa využívajú v rôznych aplikáciách, vrátane konverzačných agentov a jazykových nástrojov.", + "OpenAI API": "OpenAI API je rozhranie aplikačného programovania, ktoré umožňuje vývojárom integrovať pokročilé jazykové modely do svojich aplikácií.", + "OpenAI API Config": "Konfigurácia API OpenAI", + "OpenAI API Key is required.": "Je vyžadovaný kľúč OpenAI API.", + "OpenAI API settings updated": "", + "OpenAI URL/Key required.": "Je vyžadovaný odkaz/adresa URL alebo kľúč OpenAI.", + "or": "alebo", + "Organize your users": "", + "Other": "Iné", + "OUTPUT": "VÝSTUP", + "Output format": "Formát výstupu", + "Overview": "Prehľad", + "page": "stránka", + "Password": "Heslo", + "Paste Large Text as File": "", + "PDF document (.pdf)": "PDF dokument (.pdf)", + "PDF Extract Images (OCR)": "Extrahovanie obrázkov z PDF (OCR)", + "pending": "čaká na vybavenie", + "Permission denied when accessing media devices": "Odmietnutie povolenia pri prístupe k mediálnym zariadeniam", + "Permission denied when accessing microphone": "Prístup k mikrofónu bol zamietnutý", + "Permission denied when accessing microphone: {{error}}": "Oprávnenie zamietnuté pri prístupe k mikrofónu: {{error}}", + "Permissions": "", + "Personalization": "Personalizácia", + "Pin": "", + "Pinned": "", + "Pioneer insights": "", + "Pipeline deleted successfully": "Pipeline bola úspešne odstránená", + "Pipeline downloaded successfully": "Kanál bol úspešne stiahnutý", + "Pipelines": "", + "Pipelines Not Detected": "Prenosové kanály neboli detekované", + "Pipelines Valves": "", + "Plain text (.txt)": "Čistý text (.txt)", + "Playground": "", + "Please carefully review the following warnings:": "Prosím, pozorne si prečítajte nasledujúce upozornenia:", + "Please enter a prompt": "Prosím, zadajte zadanie.", + "Please fill in all fields.": "Prosím, vyplňte všetky polia.", + "Please select a model first.": "", + "Please select a reason": "Prosím vyberte dôvod", + "Port": "", + "Positive attitude": "Pozitívny prístup", + "Prefix ID": "", + "Prefix ID is used to avoid conflicts with other connections by adding a prefix to the model IDs - leave empty to disable": "", + "Previous 30 days": "Predchádzajúcich 30 dní", + "Previous 7 days": "Predchádzajúcich 7 dní", + "Profile Image": "Profilový obrázok", + "Prompt (e.g. Tell me a fun fact about the Roman Empire)": "Prompt (napr. Povedz mi zábavnú skutočnosť o Rímskej ríši)", + "Prompt Content": "Obsah promptu", + "Prompt created successfully": "", + "Prompt suggestions": "Návrhy výziev", + "Prompt updated successfully": "", + "Prompts": "Prompty", + "Prompts Access": "", + "Proxy URL": "", + "Pull \"{{searchValue}}\" from Ollama.com": "Stiahnite \"{{searchValue}}\" z Ollama.com", + "Pull a model from Ollama.com": "Stiahnite model z Ollama.com", + "Query Generation Prompt": "", + "Query Params": "Parametre dotazu", + "RAG Template": "Šablóna RAG", + "Rating": "Hodnotenie", + "Re-rank models by topic similarity": "Znova zoradiť modely podľa podobnosti tém.", + "Read Aloud": "Čítať nahlas", + "Record voice": "Nahrať hlas", + "Redirecting you to OpenWebUI Community": "Presmerovanie na komunitu OpenWebUI", + "Reduces the probability of generating nonsense. A higher value (e.g. 100) will give more diverse answers, while a lower value (e.g. 10) will be more conservative. (Default: 40)": "", + "Refer to yourself as \"User\" (e.g., \"User is learning Spanish\")": "Odkazujte na seba ako na \"užívateľa\" (napr. \"Užívateľ sa učí španielsky\").", + "References from": "Referencie z", + "Refused when it shouldn't have": "Odmietnuté, keď nemalo byť.", + "Regenerate": "Regenerovať", + "Release Notes": "Záznamy o vydaní", + "Relevance": "Relevancia", + "Remove": "Odstrániť", + "Remove Model": "Odstrániť model", + "Rename": "Premenovať", + "Reorder Models": "", + "Repeat Last N": "Opakovať posledných N", + "Request Mode": "Režim žiadosti", + "Reranking Model": "Model na prehodnotenie poradia", + "Reranking model disabled": "Model na prehodnotenie poradia je deaktivovaný", + "Reranking model set to \"{{reranking_model}}\"": "Model na prehodnotenie poradia nastavený na \"{{reranking_model}}\"", + "Reset": "režim Reset", + "Reset All Models": "", + "Reset Upload Directory": "Resetovať adresár nahrávania", + "Reset Vector Storage/Knowledge": "Resetovanie úložiska vektorov/znalostí", + "Response notifications cannot be activated as the website permissions have been denied. Please visit your browser settings to grant the necessary access.": "Oznámenia o odpovediach nie je možné aktivovať, pretože povolenia webu boli zamietnuté. Navštívte nastavenia svojho prehliadača a povoľte potrebný prístup.", + "Response splitting": "Rozdelenie odpovede", + "Result": "Výsledok", + "Retrieval Query Generation": "", + "Rich Text Input for Chat": "Vstup pre chat vo formáte Rich Text", + "RK": "RK", + "Role": "Funkcia", + "Rosé Pine": "Rosé Pine", + "Rosé Pine Dawn": "Rosé Pine Dawn", + "RTL": "RTL", + "Run": "Spustiť", + "Running": "Spúšťanie", + "Save": "Uložiť", + "Save & Create": "Uložiť a Vytvoriť", + "Save & Update": "Uložiť a aktualizovať", + "Save As Copy": "Uložiť ako kópiu", + "Save Tag": "Uložiť štítok", + "Saved": "Uložené", + "Saving chat logs directly to your browser's storage is no longer supported. Please take a moment to download and delete your chat logs by clicking the button below. Don't worry, you can easily re-import your chat logs to the backend through": "Ukladanie záznamov chatu priamo do úložiska vášho prehliadača už nie je podporované. Venujte prosím chvíľu stiahnutiu a vymazaniu svojich záznamov chatu kliknutím na tlačidlo nižšie. Nemajte obavy, môžete ľahko znovu importovať svoje záznamy chatu na backend prostredníctvom", + "Scroll to bottom when switching between branches": "Prejsť na koniec pri prepínaní medzi vetvami.", + "Search": "Vyhľadávanie", + "Search a model": "Vyhľadať model", + "Search Base": "", + "Search Chats": "Vyhľadávanie v chate", + "Search Collection": "Hľadať kolekciu", + "Search Filters": "", + "search for tags": "hľadanie značiek", + "Search Functions": "Vyhľadávacie funkcie", + "Search Knowledge": "Vyhľadávanie znalostí", + "Search Models": "Vyhľadávacie modely", + "Search options": "", + "Search Prompts": "Vyhľadávacie dotazy", + "Search Result Count": "Počet výsledkov hľadania", + "Search the web": "", + "Search Tools": "Nástroje na vyhľadávanie", + "SearchApi API Key": "Kľúč API pre SearchApi", + "SearchApi Engine": "Vyhľadávací engine API", + "Searched {{count}} sites_one": "Prehľadané {{count}} stránky_one", + "Searched {{count}} sites_few": "", + "Searched {{count}} sites_many": "", + "Searched {{count}} sites_other": "Prehľadané {{count}} stránky_iné", + "Searching \"{{searchQuery}}\"": "Hľadanie \"{{searchQuery}}\"", + "Searching Knowledge for \"{{searchQuery}}\"": "Vyhľadávanie znalostí pre \"{{searchQuery}}\"", + "Searxng Query URL": "Adresa URL dotazu Searxng", + "See readme.md for instructions": "Pozrite si {{readme.md}} pre pokyny.", + "See what's new": "Pozrite sa, čo je nové", + "Seed": "Semienko", + "Select a base model": "Vyberte základný model", + "Select a engine": "Vyberte engine", + "Select a function": "Vyberte funkciu", + "Select a group": "", + "Select a model": "Vyberte model", + "Select a pipeline": "Vyberte pipeline", + "Select a pipeline url": "Vyberte URL adresu kanála", + "Select a tool": "Vyberte nástroj", + "Select Engine": "Vyberte engine", + "Select Knowledge": "Vybrať znalosti", + "Select model": "Vyberte model", + "Select only one model to call": "Vyberte iba jeden model, ktorý chcete použiť", + "Selected model(s) do not support image inputs": "Vybraný(é) model(y) nepodporujú vstupy v podobe obrázkov.", + "Semantic distance to query": "Sémantická vzdialenosť k dotazu", + "Send": "Odoslať", + "Send a Message": "Odoslať správu", + "Send message": "Odoslať správu", + "Sends `stream_options: { include_usage: true }` in the request.\nSupported providers will return token usage information in the response when set.": "Odošle `stream_options: { include_usage: true }` v žiadosti. Podporovaní poskytovatelia vrátia informácie o využití tokenov v odpovedi, keď je táto možnosť nastavená.", + "September": "September", + "Serper API Key": "Kľúč API pre Serper", + "Serply API Key": "Serply API kľúč", + "Serpstack API Key": "Kľúč API pre Serpstack", + "Server connection verified": "Pripojenie k serveru overené", + "Set as default": "Nastaviť ako predvolené", + "Set CFG Scale": "Nastavte hodnotu CFG Scale", + "Set Default Model": "Nastavenie predvoleného modelu", + "Set embedding model": "", + "Set embedding model (e.g. {{model}})": "Nastavte model vkladania (napr. {{model}})", + "Set Image Size": "Nastavenie veľkosti obrázku", + "Set reranking model (e.g. {{model}})": "Nastavte model na prehodnotenie (napr. {{model}})", + "Set Sampler": "Nastavenie vzorkovača", + "Set Scheduler": "Nastavenie plánovača", + "Set Steps": "Nastavenie krokov", + "Set Task Model": "Nastaviť model úlohy", + "Set the number of GPU devices used for computation. This option controls how many GPU devices (if available) are used to process incoming requests. Increasing this value can significantly improve performance for models that are optimized for GPU acceleration but may also consume more power and GPU resources.": "", + "Set the number of worker threads used for computation. This option controls how many threads are used to process incoming requests concurrently. Increasing this value can improve performance under high concurrency workloads but may also consume more CPU resources.": "", + "Set Voice": "Nastaviť hlas", + "Set whisper model": "Nastaviť model whisper", + "Sets how far back for the model to look back to prevent repetition. (Default: 64, 0 = disabled, -1 = num_ctx)": "", + "Sets how strongly to penalize repetitions. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. (Default: 1.1)": "", + "Sets the random number seed to use for generation. Setting this to a specific number will make the model generate the same text for the same prompt. (Default: random)": "", + "Sets the size of the context window used to generate the next token. (Default: 2048)": "", + "Sets the stop sequences to use. When this pattern is encountered, the LLM will stop generating text and return. Multiple stop patterns may be set by specifying multiple separate stop parameters in a modelfile.": "", + "Settings": "Nastavenia", + "Settings saved successfully!": "Nastavenia boli úspešne uložené!", + "Share": "Zdieľať", + "Share Chat": "Zdieľať chat", + "Share to OpenWebUI Community": "Zdieľať s komunitou OpenWebUI", + "Show": "Zobraziť", + "Show \"What's New\" modal on login": "", + "Show Admin Details in Account Pending Overlay": "Zobraziť podrobnosti administrátora v prekryvnom okne s čakajúcim účtom", + "Show shortcuts": "Zobraziť klávesové skratky", + "Show your support!": "Vyjadrite svoju podporu!", + "Showcased creativity": "Predvedená kreativita", + "Sign in": "Prihlásiť sa", + "Sign in to {{WEBUI_NAME}}": "Prihlásiť sa do {{WEBUI_NAME}}", + "Sign in to {{WEBUI_NAME}} with LDAP": "", + "Sign Out": "Odhlásiť sa", + "Sign up": "Zaregistrovať sa", + "Sign up to {{WEBUI_NAME}}": "Zaregistrujte sa na {{WEBUI_NAME}}", + "Signing in to {{WEBUI_NAME}}": "Prihlasovanie do {{WEBUI_NAME}}", + "Source": "Zdroj", + "Speech Playback Speed": "Rýchlosť prehrávania reči", + "Speech recognition error: {{error}}": "Chyba rozpoznávania reči: {{error}}", + "Speech-to-Text Engine": "Motor prevodu reči na text", + "Stop": "Zastaviť", + "Stop Sequence": "Sekvencia zastavenia", + "Stream Chat Response": "Odozva chatu Stream", + "STT Model": "Model rozpoznávania reči na text (STT)", + "STT Settings": "Nastavenia STT (Rozpoznávanie reči)", + "Subtitle (e.g. about the Roman Empire)": "Titulky (napr. o Rímskej ríši)", + "Success": "Úspech", + "Successfully updated.": "Úspešne aktualizované.", + "Suggested": "Navrhované", + "Support": "Podpora", + "Support this plugin:": "Podporte tento plugin:", + "Sync directory": "Synchronizovať adresár", + "System": "Systém", + "System Instructions": "", + "System Prompt": "Systémový prompt", + "Tags Generation": "", + "Tags Generation Prompt": "Prompt na generovanie značiek", + "Tail free sampling is used to reduce the impact of less probable tokens from the output. A higher value (e.g., 2.0) will reduce the impact more, while a value of 1.0 disables this setting. (default: 1)": "", + "Tap to interrupt": "Klepnite na prerušenie", + "Tavily API Key": "Kľúč API pre Tavily", + "Tell us more:": "Povedzte nám viac.", + "Temperature": "", + "Template": "Šablóna", + "Temporary Chat": "Dočasný chat", + "Text Splitter": "Rozdeľovač textu", + "Text-to-Speech Engine": "Stroj na prevod textu na reč", + "Tfs Z": "Tfs Z", + "Thanks for your feedback!": "Ďakujeme za vašu spätnú väzbu!", + "The Application Account DN you bind with for search": "", + "The base to search for users": "", + "The batch size determines how many text requests are processed together at once. A higher batch size can increase the performance and speed of the model, but it also requires more memory. (Default: 512)": "", + "The developers behind this plugin are passionate volunteers from the community. If you find this plugin helpful, please consider contributing to its development.": "Vývojári stojaci za týmto pluginom sú zapálení dobrovoľníci z komunity. Ak považujete tento plugin za užitočný, zvážte príspevok na jeho vývoj.", + "The evaluation leaderboard is based on the Elo rating system and is updated in real-time.": "Hodnotiaca tabuľka je založená na systéme hodnotenia Elo a aktualizuje sa v reálnom čase.", + "The LDAP attribute that maps to the username that users use to sign in.": "", + "The leaderboard is currently in beta, and we may adjust the rating calculations as we refine the algorithm.": "Hodnotiaca tabuľka je momentálne v beta verzii a môžeme upraviť výpočty hodnotenia, ako budeme zdokonaľovať algoritmus.", + "The maximum file size in MB. If the file size exceeds this limit, the file will not be uploaded.": "Maximálna veľkosť súboru v MB. Ak veľkosť súboru presiahne tento limit, súbor nebude nahraný.", + "The maximum number of files that can be used at once in chat. If the number of files exceeds this limit, the files will not be uploaded.": "Maximálny počet súborov, ktoré je možné použiť naraz v chate. Ak počet súborov presiahne tento limit, súbory nebudú nahrané.", + "The score should be a value between 0.0 (0%) and 1.0 (100%).": "Skóre by malo byť hodnotou medzi 0,0 (0%) a 1,0 (100%).", + "The temperature of the model. Increasing the temperature will make the model answer more creatively. (Default: 0.8)": "", + "Theme": "Téma", + "Thinking...": "Premýšľam...", + "This action cannot be undone. Do you wish to continue?": "Túto akciu nie je možné vrátiť späť. Prajete si pokračovať?", + "This ensures that your valuable conversations are securely saved to your backend database. Thank you!": "Týmto je zaistené, že vaše cenné konverzácie sú bezpečne uložené vo vašej backendovej databáze. Ďakujeme!", + "This is an experimental feature, it may not function as expected and is subject to change at any time.": "Toto je experimentálna funkcia, nemusí fungovať podľa očakávania a môže byť kedykoľvek zmenená.", + "This option controls how many tokens are preserved when refreshing the context. For example, if set to 2, the last 2 tokens of the conversation context will be retained. Preserving context can help maintain the continuity of a conversation, but it may reduce the ability to respond to new topics. (Default: 24)": "", + "This option sets the maximum number of tokens the model can generate in its response. Increasing this limit allows the model to provide longer answers, but it may also increase the likelihood of unhelpful or irrelevant content being generated. (Default: 128)": "", + "This option will delete all existing files in the collection and replace them with newly uploaded files.": "Táto voľba odstráni všetky existujúce súbory v kolekcii a nahradí ich novo nahranými súbormi.", + "This response was generated by \"{{model}}\"": "Táto odpoveď bola vygenerovaná pomocou \"{{model}}\"", + "This will delete": "Toto odstráni", + "This will delete {{NAME}} and all its contents.": "Týmto dôjde k odstráneniu {{NAME}} a všetkých jeho obsahov.", + "This will delete all models including custom models": "", + "This will delete all models including custom models and cannot be undone.": "", + "This will reset the knowledge base and sync all files. Do you wish to continue?": "Toto obnoví znalostnú databázu a synchronizuje všetky súbory. Prajete si pokračovať?", + "Thorough explanation": "Obsiahle vysvetlenie", + "Tika": "Tika", + "Tika Server URL required.": "Je vyžadovaná URL adresa servera Tika.", + "Tiktoken": "Tiktoken", + "Tip: Update multiple variable slots consecutively by pressing the tab key in the chat input after each replacement.": "Tip: Aktualizujte postupne viacero premenných slotov stlačením klávesy Tab v chate po každej náhrade.", + "Title": "Názov", + "Title (e.g. Tell me a fun fact)": "Názov (napr. Povedz mi zaujímavosť)", + "Title Auto-Generation": "Automatické generovanie názvu", + "Title cannot be an empty string.": "Názov nemôže byť prázdny reťazec.", + "Title Generation Prompt": "Generovanie názvu promptu", + "TLS": "", + "To access the available model names for downloading,": "Pre získanie dostupných názvov modelov na stiahnutie,", + "To access the GGUF models available for downloading,": "Pre prístup k modelom GGUF dostupným na stiahnutie,", + "To access the WebUI, please reach out to the administrator. Admins can manage user statuses from the Admin Panel.": "Pre prístup k WebUI sa prosím obráťte na administrátora. Administrátori môžu spravovať stavy používateľov z Admin Panelu.", + "To attach knowledge base here, add them to the \"Knowledge\" workspace first.": "Ak chcete tu pripojiť znalostnú databázu, najprv ju pridajte do pracovného priestoru \"Knowledge\".", + "To protect your privacy, only ratings, model IDs, tags, and metadata are shared from your feedback—your chat logs remain private and are not included.": "Na ochranu vášho súkromia sú z vašej spätnej väzby zdieľané iba hodnotenia, ID modelov, značky a metadáta – vaše záznamy chatu zostávajú súkromné a nie sú zahrnuté.", + "To select actions here, add them to the \"Functions\" workspace first.": "Ak chcete tu vybrať akcie, najprv ich pridajte do pracovného priestoru \"Functions\".", + "To select filters here, add them to the \"Functions\" workspace first.": "Ak chcete tu vybrať filtre, najprv ich pridajte do pracovného priestoru „Functions“.", + "To select toolkits here, add them to the \"Tools\" workspace first.": "Ak chcete tu vybrať nástroje, pridajte ich najprv do pracovného priestoru \"Tools\".", + "Toast notifications for new updates": "Oznámenia vo forme toastov pre nové aktualizácie", + "Today": "Dnes", + "Toggle settings": "Prepnúť nastavenia", + "Toggle sidebar": "Prepnúť bočný panel", + "Token": "Token", + "Tokens To Keep On Context Refresh (num_keep)": "Tokeny, ktoré si ponechať pri obnovení kontextu (num_keep)", + "Too verbose": "Príliš rozvláčne", + "Tool created successfully": "Nástroj bol úspešne vytvorený.", + "Tool deleted successfully": "Nástroj bol úspešne odstránený.", + "Tool Description": "", + "Tool ID": "ID nástroja", + "Tool imported successfully": "Nástroj bol úspešne importovaný", + "Tool Name": "", + "Tool updated successfully": "Nástroj bol úspešne aktualizovaný.", + "Tools": "Nástroje", + "Tools Access": "", + "Tools are a function calling system with arbitrary code execution": "Nástroje sú systémom na volanie funkcií s vykonávaním ľubovoľného kódu.", + "Tools have a function calling system that allows arbitrary code execution": "Nástroje majú systém volania funkcií, ktorý umožňuje ľubovoľné spúšťanie kódu.", + "Tools have a function calling system that allows arbitrary code execution.": "Nástroje majú systém volania funkcií, ktorý umožňuje spúšťanie ľubovoľného kódu.", + "Top K": "Top K", + "Top P": "Top P", + "Transformers": "", + "Trouble accessing Ollama?": "Máte problémy s prístupom k Ollama?", + "TTS Model": "Model prevodu textu na reč (TTS)", + "TTS Settings": "Nastavenia TTS (Text-to-Speech)", + "TTS Voice": "TTS hlas", + "Type": "Napíšte", + "Type Hugging Face Resolve (Download) URL": "Zadajte URL na úspešné stiahnutie z Hugging Face.", + "Uh-oh! There was an issue connecting to {{provider}}.": "Ups! Vyskytol sa problém s pripojením k poskytovateľovi {{provider}}.", + "UI": "UI", + "Unarchive All": "Odzálohovať všetky", + "Unarchive All Archived Chats": "", + "Unarchive Chat": "", + "Unlock mysteries": "", + "Unpin": "Odopnúť", + "Unravel secrets": "", + "Untagged": "Nebola označená", + "Update": "Aktualizovať", + "Update and Copy Link": "Aktualizovať a skopírovať odkaz", + "Update for the latest features and improvements.": "Aktualizácia pre najnovšie funkcie a vylepšenia.", + "Update password": "Aktualizovať heslo", + "Updated": "Aktualizované", + "Updated at": "Aktualizované dňa", + "Updated At": "Aktualizované dňa", + "Upload": "Nahrať", + "Upload a GGUF model": "Nahrať model vo formáte GGUF", + "Upload directory": "Nahrať adresár", + "Upload files": "Nahrať súbory", + "Upload Files": "Nahrať súbory", + "Upload Pipeline": "Nahrať pipeline", + "Upload Progress": "Priebeh nahrávania", + "URL": "", + "URL Mode": "Režim URL", + "Use '#' in the prompt input to load and include your knowledge.": "Použite '#' vo vstupe promptu na načítanie a zahrnutie vašich vedomostí.", + "Use Gravatar": "Použiť Gravatar", + "Use groups to group your users and assign permissions.": "", + "Use Initials": "Použiť iniciály", + "use_mlock (Ollama)": "use_mlock (Ollama)", + "use_mmap (Ollama)": "use_mmap (Ollama)", + "user": "používateľ", + "User": "Používateľ", + "User location successfully retrieved.": "Umiestnenie používateľa bolo úspešne získané.", + "Username": "Používateľské meno", + "Users": "Používatelia", + "Using the default arena model with all models. Click the plus button to add custom models.": "Používanie predvoleného modelu arény so všetkými modelmi. Kliknutím na tlačidlo plus pridajte vlastné modely.", + "Utilize": "Využiť", + "Valid time units:": "Platné časové jednotky:", + "Valves": "Ventily", + "Valves updated": "Ventily aktualizované", + "Valves updated successfully": "Ventily boli úspešne aktualizované.", + "variable": "premenná", + "variable to have them replaced with clipboard content.": "premennú, aby bol ich obsah nahradený obsahom schránky.", + "Version": "Verzia", + "Version {{selectedVersion}} of {{totalVersions}}": "Verzia {{selectedVersion}} z {{totalVersions}}", + "Visibility": "Viditeľnosť", + "Voice": "Hlas", + "Voice Input": "Hlasový vstup", + "Warning": "Varovanie", + "Warning:": "Upozornenie:", + "Warning: Enabling this will allow users to upload arbitrary code on the server.": "", + "Warning: If you update or change your embedding model, you will need to re-import all documents.": "Varovanie: Ak aktualizujete alebo zmeníte svoj model vkladania, budete musieť všetky dokumenty znovu importovať.", + "Web": "Web", + "Web API": "Webové API", + "Web Loader Settings": "Nastavenia Web Loaderu", + "Web Search": "Vyhľadávanie na webe", + "Web Search Engine": "Webový vyhľadávač", + "Web Search Query Generation": "", + "Webhook URL": "Webhook URL", + "WebUI Settings": "Nastavenia WebUI", + "WebUI will make requests to \"{{url}}/api/chat\"": "", + "WebUI will make requests to \"{{url}}/chat/completions\"": "", + "What are you trying to achieve?": "", + "What are you working on?": "", + "What’s New in": "Čo je nové v", + "When enabled, the model will respond to each chat message in real-time, generating a response as soon as the user sends a message. This mode is useful for live chat applications, but may impact performance on slower hardware.": "", + "wherever you are": "kdekoľvek ste", + "Whisper (Local)": "Whisper (Lokálne)", + "Why?": "Prečo?", + "Widescreen Mode": "Režim širokouhlého zobrazenia", + "Won": "Vyhral", + "Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. (Default: 0.9)": "", + "Workspace": "", + "Workspace Permissions": "", + "Write a prompt suggestion (e.g. Who are you?)": "Navrhnite otázku (napr. Kto ste?)", + "Write a summary in 50 words that summarizes [topic or keyword].": "Napíšte zhrnutie na 50 slov, ktoré zhrňuje [tému alebo kľúčové slovo].", + "Write something...": "Napíšte niečo...", + "Write your model template content here": "", + "Yesterday": "Včera", + "You": "Vy", + "You can only chat with a maximum of {{maxCount}} file(s) at a time.": "Môžete komunikovať len s maximálne {{maxCount}} súbor(ami) naraz.", + "You can personalize your interactions with LLMs by adding memories through the 'Manage' button below, making them more helpful and tailored to you.": "Môžete personalizovať svoje interakcie s LLM pridaním spomienok prostredníctvom tlačidla 'Spravovať' nižšie, čo ich urobí pre vás užitočnejšími a lepšie prispôsobenými.", + "You cannot upload an empty file.": "Nemôžete nahrať prázdny súbor.", + "You do not have permission to upload files.": "", + "You have no archived conversations.": "Nemáte žiadne archivované konverzácie.", + "You have shared this chat": "Zdieľali ste tento chat.", + "You're a helpful assistant.": "Ste užitočný asistent.", + "You're now logged in.": "Teraz ste prihlásený(-á).", + "Your account status is currently pending activation.": "Stav vášho účtu je aktuálne čakajúci na aktiváciu.", + "Your entire contribution will go directly to the plugin developer; Open WebUI does not take any percentage. However, the chosen funding platform might have its own fees.": "Celý váš príspevok pôjde priamo vývojárovi pluginu; Open WebUI si neberie žiadne percento. Zvolená platforma na financovanie však môže mať vlastné poplatky.", + "Youtube": "YouTube", + "Youtube Loader Settings": "Nastavenia YouTube loaderu" +} diff --git a/src/lib/i18n/locales/sr-RS/translation.json b/src/lib/i18n/locales/sr-RS/translation.json index f5cd6e73a..3baa948eb 100644 --- a/src/lib/i18n/locales/sr-RS/translation.json +++ b/src/lib/i18n/locales/sr-RS/translation.json @@ -1,7 +1,7 @@ { - "-1 for no limit, or a positive integer for a specific limit": "", + "-1 for no limit, or a positive integer for a specific limit": "-1 за бесконачно или позитивни број за одређено ограничење", "'s', 'm', 'h', 'd', 'w' or '-1' for no expiration.": "„s“, „m“, „h“, „d“, „w“ или „-1“ за без истека.", - "(e.g. `sh webui.sh --api --api-auth username_password`)": "", + "(e.g. `sh webui.sh --api --api-auth username_password`)": "(нпр. `sh webui.sh --api --api-auth username_password`)", "(e.g. `sh webui.sh --api`)": "(нпр. `sh webui.sh --api`)", "(latest)": "(најновије)", "{{ models }}": "{{ модели }}", @@ -12,58 +12,58 @@ "A task model is used when performing tasks such as generating titles for chats and web search queries": "Модел задатка се користи приликом извршавања задатака као што су генерисање наслова за ћаскања и упите за Веб претрагу", "a user": "корисник", "About": "О нама", - "Access": "", - "Access Control": "", - "Accessible to all users": "", + "Access": "Приступ", + "Access Control": "Контрола приступа", + "Accessible to all users": "Доступно свим корисницима", "Account": "Налог", - "Account Activation Pending": "", + "Account Activation Pending": "Налози за активирање", "Accurate information": "Прецизне информације", - "Actions": "", - "Activate this command by typing \"/{{COMMAND}}\" to chat input.": "", - "Active Users": "", + "Actions": "Радње", + "Activate this command by typing \"/{{COMMAND}}\" to chat input.": "Покрените ову наредбу куцањем \"/{{COMMAND}}\" у ћаскање.", + "Active Users": "Активни корисници", "Add": "Додај", - "Add a model ID": "", + "Add a model ID": "Додај ИБ модела", "Add a short description about what this model does": "Додавање кратког описа о томе шта овај модел ради", "Add a tag": "Додај ознаку", - "Add Arena Model": "", - "Add Connection": "", - "Add Content": "", - "Add content here": "", + "Add Arena Model": "Додај модел Арене", + "Add Connection": "Додај везу", + "Add Content": "Додај садржај", + "Add content here": "Додај садржај овде", "Add custom prompt": "Додај прилагођен упит", "Add Files": "Додај датотеке", - "Add Group": "", + "Add Group": "Додај групу", "Add Memory": "Додај меморију", "Add Model": "Додај модел", - "Add Tag": "", + "Add Tag": "Додај ознаку", "Add Tags": "Додај ознаке", - "Add text content": "", + "Add text content": "Додај садржај текста", "Add User": "Додај корисника", - "Add User Group": "", + "Add User Group": "Додај корисничку групу", "Adjusting these settings will apply changes universally to all users.": "Прилагођавање ових подешавања ће применити промене на све кориснике.", "admin": "админ", - "Admin": "", + "Admin": "Админ", "Admin Panel": "Админ табла", "Admin Settings": "Админ подешавања", - "Admins have access to all tools at all times; users need tools assigned per model in the workspace.": "", + "Admins have access to all tools at all times; users need tools assigned per model in the workspace.": "Админи имају приступ свим алатима у сваком тренутку, корисницима је потребно доделити алате по моделу у радном простору", "Advanced Parameters": "Напредни параметри", "Advanced Params": "Напредни парамови", - "All chats": "", + "All chats": "Сва ћаскања", "All Documents": "Сви документи", - "All models deleted successfully": "", - "Allow Chat Delete": "", + "All models deleted successfully": "Сви модели су успешно обрисани", + "Allow Chat Delete": "Дозволи брисање ћаскања", "Allow Chat Deletion": "Дозволи брисање ћаскања", - "Allow Chat Edit": "", - "Allow File Upload": "", - "Allow non-local voices": "", - "Allow Temporary Chat": "", - "Allow User Location": "", - "Allow Voice Interruption in Call": "", + "Allow Chat Edit": "Дозволи измену ћаскања", + "Allow File Upload": "Дозволи отпремање датотека", + "Allow non-local voices": "Дозволи нелокалне гласове", + "Allow Temporary Chat": "Дозволи привремена ћаскања", + "Allow User Location": "Дозволи корисничку локацију", + "Allow Voice Interruption in Call": "Дозволи прекид гласа у позиву", "Already have an account?": "Већ имате налог?", "Alternative to the top_p, and aims to ensure a balance of quality and variety. The parameter p represents the minimum probability for a token to be considered, relative to the probability of the most likely token. For example, with p=0.05 and the most likely token having a probability of 0.9, logits with a value less than 0.045 are filtered out. (Default: 0.0)": "", - "Amazing": "", + "Amazing": "Невероватно", "an assistant": "помоћник", "and": "и", - "and {{COUNT}} more": "", + "and {{COUNT}} more": "и још {{COUNT}}", "and create a new shared link.": "и направи нову дељену везу.", "API Base URL": "Основна адреса API-ја", "API Key": "API кључ", @@ -73,35 +73,35 @@ "Application DN Password": "", "applies to all users with the \"user\" role": "", "April": "Април", - "Archive": "Архива", - "Archive All Chats": "Архивирај све ћаскања", - "Archived Chats": "Архивирана ћаскања", + "Archive": "Архивирај", + "Archive All Chats": "Архивирај сва ћаскања", + "Archived Chats": "Архиве", "archived-chat-export": "", "Are you sure you want to unarchive all archived chats?": "", "Are you sure?": "Да ли сте сигурни?", - "Arena Models": "", - "Artifacts": "", - "Ask a question": "", - "Assistant": "", + "Arena Models": "Модели са Арене", + "Artifacts": "Артефакти", + "Ask a question": "Постави питање", + "Assistant": "Помоћник", "Attach file": "Приложи датотеку", "Attention to detail": "Пажња на детаље", - "Attribute for Username": "", + "Attribute for Username": "Особина корисника", "Audio": "Звук", "August": "Август", - "Authenticate": "", + "Authenticate": "Идентификација", "Auto-Copy Response to Clipboard": "Самостално копирање одговора у оставу", "Auto-playback response": "Самостално пуштање одговора", - "Autocomplete Generation": "", - "Autocomplete Generation Input Max Length": "", - "Automatic1111": "", - "AUTOMATIC1111 Api Auth String": "", + "Autocomplete Generation": "Стварање самодовршавања", + "Autocomplete Generation Input Max Length": "Најдужи улаз стварања самодовршавања", + "Automatic1111": "Automatic1111", + "AUTOMATIC1111 Api Auth String": "Automatic1111 Api ниска идентификације", "AUTOMATIC1111 Base URL": "Основна адреса за AUTOMATIC1111", "AUTOMATIC1111 Base URL is required.": "Потребна је основна адреса за AUTOMATIC1111.", - "Available list": "", + "Available list": "Списак доступног", "available!": "доступно!", - "Awful": "", - "Azure AI Speech": "", - "Azure Region": "", + "Awful": "Грозно", + "Azure AI Speech": "Azure AI говор", + "Azure Region": "Azure област", "Back": "Назад", "Bad Response": "Лош одговор", "Banners": "Барјаке", @@ -114,14 +114,14 @@ "Brave Search API Key": "Апи кључ за храбру претрагу", "By {{name}}": "", "Bypass SSL verification for Websites": "Заобиђи SSL потврђивање за веб странице", - "Call": "", + "Call": "Позив", "Call feature is not supported when using Web STT engine": "", - "Camera": "", + "Camera": "Камера", "Cancel": "Откажи", "Capabilities": "Могућности", "Certificate Path": "", "Change Password": "Промени лозинку", - "Character": "", + "Character": "Знак", "Character limit for autocomplete generation input": "", "Chart new frontiers": "", "Chat": "Ћаскање", @@ -129,8 +129,8 @@ "Chat Bubble UI": "Интерфејс балона ћаскања", "Chat Controls": "", "Chat direction": "Смер ћаскања", - "Chat Overview": "", - "Chat Permissions": "", + "Chat Overview": "Преглед ћаскања", + "Chat Permissions": "Дозволе ћаскања", "Chat Tags Auto-Generation": "", "Chats": "Ћаскања", "Check Again": "Провери поново", @@ -156,7 +156,7 @@ "click here.": "кликните овде.", "Click on the user role button to change a user's role.": "Кликните на дугме за улогу корисника да промените улогу корисника.", "Clipboard write permission denied. Please check your browser settings to grant the necessary access.": "", - "Clone": "Клон", + "Clone": "Клонирај", "Close": "Затвори", "Code execution": "", "Code formatted successfully": "", @@ -185,29 +185,29 @@ "Continue with Email": "", "Continue with LDAP": "", "Control how message text is split for TTS requests. 'Punctuation' splits into sentences, 'paragraphs' splits into paragraphs, and 'none' keeps the message as a single string.": "", - "Controls": "", + "Controls": "Контроле", "Controls the balance between coherence and diversity of the output. A lower value will result in more focused and coherent text. (Default: 5.0)": "", - "Copied": "", + "Copied": "Копирано", "Copied shared chat URL to clipboard!": "Адреса дељеног ћаскања ископирана у оставу!", - "Copied to clipboard": "", + "Copied to clipboard": "Копирано у оставу", "Copy": "Копирај", "Copy last code block": "Копирај последњи блок кода", "Copy last response": "Копирај последњи одговор", "Copy Link": "Копирај везу", "Copy to clipboard": "", "Copying to clipboard was successful!": "Успешно копирање у оставу!", - "Create": "", - "Create a knowledge base": "", + "Create": "Направи", + "Create a knowledge base": "Направи базу знања", "Create a model": "Креирање модела", "Create Account": "Направи налог", - "Create Admin Account": "", - "Create Group": "", - "Create Knowledge": "", + "Create Admin Account": "Направи админ налог", + "Create Group": "Направи групу", + "Create Knowledge": "Направи знање", "Create new key": "Направи нови кључ", "Create new secret key": "Направи нови тајни кључ", "Created at": "Направљено у", "Created At": "Направљено у", - "Created by": "", + "Created by": "Направио/ла", "CSV Import": "", "Current Model": "Тренутни модел", "Current Password": "Тренутна лозинка", @@ -220,8 +220,8 @@ "Default (SentenceTransformers)": "Подразумевано (SentenceTransformers)", "Default Model": "Подразумевани модел", "Default model updated": "Подразумевани модел ажуриран", - "Default Models": "", - "Default permissions": "", + "Default Models": "Подразумевани модели", + "Default permissions": "Подразумевана овлашћења", "Default permissions updated successfully": "", "Default Prompt Suggestions": "Подразумевани предлози упита", "Default to 389 or 636 if TLS is enabled": "", @@ -247,14 +247,14 @@ "Description": "Опис", "Didn't fully follow instructions": "Упутства нису праћена у потпуности", "Disabled": "", - "Discover a function": "", + "Discover a function": "Откријте функцију", "Discover a model": "Откријте модел", "Discover a prompt": "Откриј упит", - "Discover a tool": "", + "Discover a tool": "Откријте алат", "Discover wonders": "", - "Discover, download, and explore custom functions": "", + "Discover, download, and explore custom functions": "Откријте, преузмите и истражите прилагођене функције", "Discover, download, and explore custom prompts": "Откријте, преузмите и истражите прилагођене упите", - "Discover, download, and explore custom tools": "", + "Discover, download, and explore custom tools": "Откријте, преузмите и истражите прилагођене алате", "Discover, download, and explore model presets": "Откријте, преузмите и истражите образце модела", "Dismissible": "", "Display": "", @@ -350,7 +350,7 @@ "Enter server label": "", "Enter server port": "", "Enter stop sequence": "Унесите секвенцу заустављања", - "Enter system prompt": "", + "Enter system prompt": "Унеси системски упит", "Enter Tavily API Key": "", "Enter Tika Server URL": "", "Enter Top K": "Унесите Топ К", @@ -363,8 +363,8 @@ "Enter Your Role": "Унесите вашу улогу", "Enter Your Username": "", "Error": "Грешка", - "ERROR": "", - "Evaluations": "", + "ERROR": "ГРЕШКА", + "Evaluations": "Процењивања", "Example: (&(objectClass=inetOrgPerson)(uid=%s))": "", "Example: ALL": "", "Example: ou=users,dc=foo,dc=example": "", @@ -378,7 +378,7 @@ "Export chat (.json)": "", "Export Chats": "Извези ћаскања", "Export Config to JSON File": "", - "Export Functions": "", + "Export Functions": "Извези функције", "Export Models": "Извези моделе", "Export Presets": "", "Export Prompts": "Извези упите", @@ -392,17 +392,17 @@ "Failed to update settings": "", "Failed to upload file.": "", "February": "Фебруар", - "Feedback History": "", - "Feedbacks": "", + "Feedback History": "Историјат повратних података", + "Feedbacks": "Повратни подаци", "Feel free to add specific details": "Слободно додајте специфичне детаље", - "File": "", + "File": "Датотека", "File added successfully.": "", "File content updated successfully.": "", "File Mode": "Режим датотеке", "File not found.": "Датотека није пронађена.", "File removed successfully.": "", "File size should not exceed {{maxSize}} MB.": "", - "Files": "", + "Files": "Датотеке", "Filter is now globally disabled": "", "Filter is now globally enabled": "", "Filters": "", @@ -427,7 +427,7 @@ "Function is now globally enabled": "", "Function Name": "", "Function updated successfully": "", - "Functions": "", + "Functions": "Функције", "Functions allow arbitrary code execution": "", "Functions allow arbitrary code execution.": "", "Functions imported successfully": "", @@ -442,12 +442,12 @@ "Good Response": "Добар одговор", "Google PSE API Key": "Гоогле ПСЕ АПИ кључ", "Google PSE Engine Id": "Гоогле ПСЕ ИД мотора", - "Group created successfully": "", - "Group deleted successfully": "", - "Group Description": "", - "Group Name": "", - "Group updated successfully": "", - "Groups": "", + "Group created successfully": "Група направљена успешно", + "Group deleted successfully": "Група обрисана успешно", + "Group Description": "Опис групе", + "Group Name": "Назив групе", + "Group updated successfully": "Група измењена успешно", + "Groups": "Групе", "h:mm a": "h:mm a", "Haptic Feedback": "", "has no conversations.": "нема разговора.", @@ -470,7 +470,7 @@ "Images": "Слике", "Import Chats": "Увези ћаскања", "Import Config from JSON File": "", - "Import Functions": "", + "Import Functions": "Увези функције", "Import Models": "Увези моделе", "Import Presets": "", "Import Prompts": "Увези упите", @@ -511,7 +511,7 @@ "Last Modified": "", "LDAP": "", "LDAP server updated": "", - "Leaderboard": "", + "Leaderboard": "Ранг листа", "Leave empty for unlimited": "", "Leave empty to include all models from \"{{URL}}/api/tags\" endpoint": "", "Leave empty to include all models from \"{{URL}}/models\" endpoint": "", @@ -522,7 +522,7 @@ "LLMs can make mistakes. Verify important information.": "ВЈМ-ови (LLM-ови) могу правити грешке. Проверите важне податке.", "Local": "", "Local Models": "", - "Lost": "", + "Lost": "Пораза", "LTR": "ЛНД", "Made by OpenWebUI Community": "Израдила OpenWebUI заједница", "Make sure to enclose them with": "Уверите се да их затворите са", @@ -556,7 +556,7 @@ "MMMM DD, YYYY": "ММММ ДД, ГГГГ", "MMMM DD, YYYY HH:mm": "ММММ ДД, ГГГГ ЧЧ:мм", "MMMM DD, YYYY hh:mm:ss A": "", - "Model": "", + "Model": "Модел", "Model '{{modelName}}' has been successfully downloaded.": "Модел „{{modelName}}“ је успешно преузет.", "Model '{{modelTag}}' is already in queue for downloading.": "Модел „{{modelTag}}“ је већ у реду за преузимање.", "Model {{modelId}} not found": "Модел {{modelId}} није пронађен", @@ -643,11 +643,11 @@ "OpenAI API settings updated": "", "OpenAI URL/Key required.": "Потребан је OpenAI URL/кључ.", "or": "или", - "Organize your users": "", + "Organize your users": "Организујте ваше кориснике", "Other": "Остало", "OUTPUT": "", - "Output format": "", - "Overview": "", + "Output format": "Формат излаза", + "Overview": "Преглед", "page": "", "Password": "Лозинка", "Paste Large Text as File": "", @@ -659,8 +659,8 @@ "Permission denied when accessing microphone: {{error}}": "Приступ микрофону је одбијен: {{error}}", "Permissions": "", "Personalization": "Прилагођавање", - "Pin": "", - "Pinned": "", + "Pin": "Закачи", + "Pinned": "Закачено", "Pioneer insights": "", "Pipeline deleted successfully": "", "Pipeline downloaded successfully": "", @@ -694,7 +694,7 @@ "Query Generation Prompt": "", "Query Params": "Параметри упита", "RAG Template": "RAG шаблон", - "Rating": "", + "Rating": "Оцена", "Re-rank models by topic similarity": "", "Read Aloud": "Прочитај наглас", "Record voice": "Сними глас", @@ -746,7 +746,7 @@ "Search Collection": "", "Search Filters": "", "search for tags": "", - "Search Functions": "", + "Search Functions": "Претражи функције", "Search Knowledge": "", "Search Models": "Модели претраге", "Search options": "", @@ -766,15 +766,15 @@ "See what's new": "Погледај шта је ново", "Seed": "Семе", "Select a base model": "Избор основног модела", - "Select a engine": "", - "Select a function": "", - "Select a group": "", + "Select a engine": "Изабери мотор", + "Select a function": "Изабери функцију", + "Select a group": "Изабери групу", "Select a model": "Изабери модел", "Select a pipeline": "Избор цевовода", "Select a pipeline url": "Избор урл адресе цевовода", - "Select a tool": "", - "Select Engine": "", - "Select Knowledge": "", + "Select a tool": "Изабери алат", + "Select Engine": "Изабери мотор", + "Select Knowledge": "Изабери знање", "Select model": "Изабери модел", "Select only one model to call": "", "Selected model(s) do not support image inputs": "Изабрани модели не подржавају уносе слика", @@ -853,7 +853,7 @@ "Tell us more:": "Реците нам више:", "Temperature": "Температура", "Template": "Шаблон", - "Temporary Chat": "", + "Temporary Chat": "Привремено ћаскање", "Text Splitter": "", "Text-to-Speech Engine": "Мотор за текст у говор", "Tfs Z": "Tfs Z", @@ -916,8 +916,8 @@ "Tool imported successfully": "", "Tool Name": "", "Tool updated successfully": "", - "Tools": "", - "Tools Access": "", + "Tools": "Алати", + "Tools Access": "Приступ алатима", "Tools are a function calling system with arbitrary code execution": "", "Tools have a function calling system that allows arbitrary code execution": "", "Tools have a function calling system that allows arbitrary code execution.": "", @@ -957,7 +957,7 @@ "URL Mode": "Режим адресе", "Use '#' in the prompt input to load and include your knowledge.": "", "Use Gravatar": "Користи Граватар", - "Use groups to group your users and assign permissions.": "", + "Use groups to group your users and assign permissions.": "Користите групе да бисте разврстали ваше кориснике и доделили овлашћења.", "Use Initials": "Користи иницијале", "use_mlock (Ollama)": "усе _млоцк (Оллама)", "use_mmap (Ollama)": "усе _ммап (Оллама)", @@ -969,9 +969,9 @@ "Using the default arena model with all models. Click the plus button to add custom models.": "", "Utilize": "Искористи", "Valid time units:": "Важеће временске јединице:", - "Valves": "", - "Valves updated": "", - "Valves updated successfully": "", + "Valves": "Вентили", + "Valves updated": "Вентили ажурирани", + "Valves updated successfully": "Вентили успешно ажурирани", "variable": "променљива", "variable to have them replaced with clipboard content.": "променљива за замену са садржајем оставе.", "Version": "Издање", @@ -1001,7 +1001,7 @@ "Whisper (Local)": "", "Why?": "", "Widescreen Mode": "", - "Won": "", + "Won": "Победа", "Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. (Default: 0.9)": "", "Workspace": "Радни простор", "Workspace Permissions": "",