0.15.7

-----BEGIN PGP SIGNATURE-----
 
 iQGzBAABCAAdFiEEFK5K98uJ0hmL9JLza6DRCN7QEf8FAmgPMZ8ACgkQa6DRCN7Q
 Ef+gkgv/Ved5ez/UgKHiGyxMM9MWiO9JDWKBrZP+NbhqSIz2B7efCg7PcEWveqCR
 ma5pD2T83z9lYsI4VYB/08HRdF4w5FbFbZXg5zy7R9OMc/5oEZ1tyxiOp/RLzOqi
 kCDhHX5CVadUjC98oy9q41S+AmlpV5hjjl5ZQHH9XIx/Uy/0LZMdhB4EOhDwNqj2
 MU0xekTEaouJFeIe1ewyVBZd2GC18EDjqv9ABiBeaJwx97SQ93pCNhjxnn2wm5cT
 Q96qkkNb/E3JoaOPhws5/pGLM+5SK0dYFTZkwWER1GUfrh+5wzLWbzAKwoewDRaV
 g62waOORvaYphXKh7KxmrC3Gb9eGMrK9haRwJfyfxzHCwqtI4+SFnv+izapU7hlm
 c24s73p2v9sFzjHBWEPqHvuJ7F5Q8odR40ECGhBmzgJVnXdDDu+h69MYWDq2LPAl
 3295Wa4zjF9krC9R9BU+ra7DLw90TA53Cf3yXdMyuoOpU41uqtj/7iF8p2axnIxL
 srqhqJWw
 =3gSn
 -----END PGP SIGNATURE-----

Merge tag '0.15.7' into e-260

0.15.7

.markdownlint.json

@@ -0,0 +1,3 @@
+{
+    "MD024": false
+}

CHANGELOG.md

@@ -5,6 +5,20 @@ All notable changes to Dify will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.15.7] - 2025-04-27
+
+### Added
+
+- Added support for GPT-4.1 in model providers (#18912)
+- Added support for Amazon Bedrock DeepSeek-R1 model (#18908)
+- Added support for Amazon Bedrock Claude Sonnet 3.7 model (#18788)
+- Refined version compatibility logic in app DSL service
+
+### Fixed
+
+- Fixed issue with creating apps from template categories (#18807, #18868)
+- Fixed DSL version check when creating apps from explore templates (#18872, #18878)
+
 ## [0.15.6] - 2025-04-22
 
 ### Security

api/configs/packaging/__init__.py

@@ -9,7 +9,7 @@ class PackagingInfo(BaseSettings):
 
     CURRENT_VERSION: str = Field(
         description="Dify version",
-        default="0.15.6",
+        default="0.15.7",
     )
 
     COMMIT_SHA: str = Field(

api/controllers/console/auth/forgot_password.py

@@ -6,13 +6,9 @@ from flask_restful import Resource, reqparse  # type: ignore
 
 from constants.languages import languages
 from controllers.console import api
-from controllers.console.auth.error import (EmailCodeError, InvalidEmailError,
+from controllers.console.auth.error import EmailCodeError, InvalidEmailError, InvalidTokenError, PasswordMismatchError
-                                            InvalidTokenError,
+from controllers.console.error import AccountInFreezeError, AccountNotFound, EmailSendIpLimitError
-                                            PasswordMismatchError)
+from controllers.console.wraps import email_password_login_enabled, setup_required
-from controllers.console.error import (AccountInFreezeError, AccountNotFound,
-                                       EmailSendIpLimitError)
-from controllers.console.wraps import (email_password_login_enabled,
-                                       setup_required)
 from events.tenant_event import tenant_was_created
 from extensions.ext_database import db
 from libs.helper import email, extract_remote_ip

api/controllers/console/wraps.py

@@ -11,8 +11,7 @@ from models.model import DifySetup
 from services.feature_service import FeatureService, LicenseStatus
 from services.operation_service import OperationService
 
-from .error import (NotInitValidateError, NotSetupError,
+from .error import NotInitValidateError, NotSetupError, UnauthorizedAndForceLogout
-                    UnauthorizedAndForceLogout)
 
 
 def account_initialization_required(view):

api/core/agent/cot_agent_runner.py

@@ -104,7 +104,6 @@ class CotAgentRunner(BaseAgentRunner, ABC):
 
             # recalc llm max tokens
             prompt_messages = self._organize_prompt_messages()
-            self.recalc_llm_max_tokens(self.model_config, prompt_messages)
             # invoke model
             chunks = model_instance.invoke_llm(
                 prompt_messages=prompt_messages,

api/core/agent/fc_agent_runner.py

@@ -84,7 +84,6 @@ class FunctionCallAgentRunner(BaseAgentRunner):
 
             # recalc llm max tokens
             prompt_messages = self._organize_prompt_messages()
-            self.recalc_llm_max_tokens(self.model_config, prompt_messages)
             # invoke model
             chunks: Union[Generator[LLMResultChunk, None, None], LLMResult] = model_instance.invoke_llm(
                 prompt_messages=prompt_messages,

api/core/app/apps/agent_chat/app_runner.py

@@ -55,20 +55,6 @@ class AgentChatAppRunner(AppRunner):
         query = application_generate_entity.query
         files = application_generate_entity.files
 
-        # Pre-calculate the number of tokens of the prompt messages,
-        # and return the rest number of tokens by model context token size limit and max token size limit.
-        # If the rest number of tokens is not enough, raise exception.
-        # Include: prompt template, inputs, query(optional), files(optional)
-        # Not Include: memory, external data, dataset context
-        self.get_pre_calculate_rest_tokens(
-            app_record=app_record,
-            model_config=application_generate_entity.model_conf,
-            prompt_template_entity=app_config.prompt_template,
-            inputs=inputs,
-            files=files,
-            query=query,
-        )
-
         memory = None
         if application_generate_entity.conversation_id:
             # get memory of conversation (read-only)

api/core/app/apps/base_app_runner.py

@@ -15,10 +15,8 @@ from core.app.features.annotation_reply.annotation_reply import AnnotationReplyF
 from core.app.features.hosting_moderation.hosting_moderation import HostingModerationFeature
 from core.external_data_tool.external_data_fetch import ExternalDataFetch
 from core.memory.token_buffer_memory import TokenBufferMemory
-from core.model_manager import ModelInstance
 from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk, LLMResultChunkDelta, LLMUsage
 from core.model_runtime.entities.message_entities import AssistantPromptMessage, PromptMessage
-from core.model_runtime.entities.model_entities import ModelPropertyKey
 from core.model_runtime.errors.invoke import InvokeBadRequestError
 from core.moderation.input_moderation import InputModeration
 from core.prompt.advanced_prompt_transform import AdvancedPromptTransform
@@ -31,106 +29,6 @@ if TYPE_CHECKING:
 
 
 class AppRunner:
-    def get_pre_calculate_rest_tokens(
-        self,
-        app_record: App,
-        model_config: ModelConfigWithCredentialsEntity,
-        prompt_template_entity: PromptTemplateEntity,
-        inputs: Mapping[str, str],
-        files: Sequence["File"],
-        query: Optional[str] = None,
-    ) -> int:
-        """
-        Get pre calculate rest tokens
-        :param app_record: app record
-        :param model_config: model config entity
-        :param prompt_template_entity: prompt template entity
-        :param inputs: inputs
-        :param files: files
-        :param query: query
-        :return:
-        """
-        # Invoke model
-        model_instance = ModelInstance(
-            provider_model_bundle=model_config.provider_model_bundle, model=model_config.model
-        )
-
-        model_context_tokens = model_config.model_schema.model_properties.get(ModelPropertyKey.CONTEXT_SIZE)
-
-        max_tokens = 0
-        for parameter_rule in model_config.model_schema.parameter_rules:
-            if parameter_rule.name == "max_tokens" or (
-                parameter_rule.use_template and parameter_rule.use_template == "max_tokens"
-            ):
-                max_tokens = (
-                    model_config.parameters.get(parameter_rule.name)
-                    or model_config.parameters.get(parameter_rule.use_template or "")
-                ) or 0
-
-        if model_context_tokens is None:
-            return -1
-
-        if max_tokens is None:
-            max_tokens = 0
-
-        # get prompt messages without memory and context
-        prompt_messages, stop = self.organize_prompt_messages(
-            app_record=app_record,
-            model_config=model_config,
-            prompt_template_entity=prompt_template_entity,
-            inputs=inputs,
-            files=files,
-            query=query,
-        )
-
-        prompt_tokens = model_instance.get_llm_num_tokens(prompt_messages)
-
-        rest_tokens: int = model_context_tokens - max_tokens - prompt_tokens
-        if rest_tokens < 0:
-            raise InvokeBadRequestError(
-                "Query or prefix prompt is too long, you can reduce the prefix prompt, "
-                "or shrink the max token, or switch to a llm with a larger token limit size."
-            )
-
-        return rest_tokens
-
-    def recalc_llm_max_tokens(
-        self, model_config: ModelConfigWithCredentialsEntity, prompt_messages: list[PromptMessage]
-    ):
-        # recalc max_tokens if sum(prompt_token +  max_tokens) over model token limit
-        model_instance = ModelInstance(
-            provider_model_bundle=model_config.provider_model_bundle, model=model_config.model
-        )
-
-        model_context_tokens = model_config.model_schema.model_properties.get(ModelPropertyKey.CONTEXT_SIZE)
-
-        max_tokens = 0
-        for parameter_rule in model_config.model_schema.parameter_rules:
-            if parameter_rule.name == "max_tokens" or (
-                parameter_rule.use_template and parameter_rule.use_template == "max_tokens"
-            ):
-                max_tokens = (
-                    model_config.parameters.get(parameter_rule.name)
-                    or model_config.parameters.get(parameter_rule.use_template or "")
-                ) or 0
-
-        if model_context_tokens is None:
-            return -1
-
-        if max_tokens is None:
-            max_tokens = 0
-
-        prompt_tokens = model_instance.get_llm_num_tokens(prompt_messages)
-
-        if prompt_tokens + max_tokens > model_context_tokens:
-            max_tokens = max(model_context_tokens - prompt_tokens, 16)
-
-            for parameter_rule in model_config.model_schema.parameter_rules:
-                if parameter_rule.name == "max_tokens" or (
-                    parameter_rule.use_template and parameter_rule.use_template == "max_tokens"
-                ):
-                    model_config.parameters[parameter_rule.name] = max_tokens
-
     def organize_prompt_messages(
         self,
         app_record: App,

api/core/app/apps/chat/app_runner.py

@@ -50,20 +50,6 @@ class ChatAppRunner(AppRunner):
         query = application_generate_entity.query
         files = application_generate_entity.files
 
-        # Pre-calculate the number of tokens of the prompt messages,
-        # and return the rest number of tokens by model context token size limit and max token size limit.
-        # If the rest number of tokens is not enough, raise exception.
-        # Include: prompt template, inputs, query(optional), files(optional)
-        # Not Include: memory, external data, dataset context
-        self.get_pre_calculate_rest_tokens(
-            app_record=app_record,
-            model_config=application_generate_entity.model_conf,
-            prompt_template_entity=app_config.prompt_template,
-            inputs=inputs,
-            files=files,
-            query=query,
-        )
-
         memory = None
         if application_generate_entity.conversation_id:
             # get memory of conversation (read-only)
@@ -194,9 +180,6 @@ class ChatAppRunner(AppRunner):
         if hosting_moderation_result:
             return
 
-        # Re-calculate the max tokens if sum(prompt_token +  max_tokens) over model token limit
-        self.recalc_llm_max_tokens(model_config=application_generate_entity.model_conf, prompt_messages=prompt_messages)
-
         # Invoke model
         model_instance = ModelInstance(
             provider_model_bundle=application_generate_entity.model_conf.provider_model_bundle,

api/core/app/apps/completion/app_runner.py

@@ -43,20 +43,6 @@ class CompletionAppRunner(AppRunner):
         query = application_generate_entity.query
         files = application_generate_entity.files
 
-        # Pre-calculate the number of tokens of the prompt messages,
-        # and return the rest number of tokens by model context token size limit and max token size limit.
-        # If the rest number of tokens is not enough, raise exception.
-        # Include: prompt template, inputs, query(optional), files(optional)
-        # Not Include: memory, external data, dataset context
-        self.get_pre_calculate_rest_tokens(
-            app_record=app_record,
-            model_config=application_generate_entity.model_conf,
-            prompt_template_entity=app_config.prompt_template,
-            inputs=inputs,
-            files=files,
-            query=query,
-        )
-
         # organize all inputs and template to prompt messages
         # Include: prompt template, inputs, query(optional), files(optional)
         prompt_messages, stop = self.organize_prompt_messages(
@@ -152,9 +138,6 @@ class CompletionAppRunner(AppRunner):
         if hosting_moderation_result:
             return
 
-        # Re-calculate the max tokens if sum(prompt_token +  max_tokens) over model token limit
-        self.recalc_llm_max_tokens(model_config=application_generate_entity.model_conf, prompt_messages=prompt_messages)
-
         # Invoke model
         model_instance = ModelInstance(
             provider_model_bundle=application_generate_entity.model_conf.provider_model_bundle,

api/core/memory/token_buffer_memory.py

@@ -26,7 +26,7 @@ class TokenBufferMemory:
         self.model_instance = model_instance
 
     def get_history_prompt_messages(
-        self, max_token_limit: int = 2000, message_limit: Optional[int] = None
+        self, max_token_limit: int = 100000, message_limit: Optional[int] = None
     ) -> Sequence[PromptMessage]:
         """
         Get history prompt messages.

api/core/model_runtime/model_providers/bedrock/llm/eu.anthropic.claude-3.7-sonnet-v1.yaml

@@ -0,0 +1,115 @@
+model: us.anthropic.claude-3-7-sonnet-20250219-v1:0
+label:
+  en_US: Claude 3.7 Sonnet(US.Cross Region Inference)
+icon: icon_s_en.svg
+model_type: llm
+features:
+  - agent-thought
+  - vision
+  - tool-call
+  - stream-tool-call
+model_properties:
+  mode: chat
+  context_size: 200000
+# docs: https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters-anthropic-claude-messages.html
+parameter_rules:
+  - name: enable_cache
+    label:
+      zh_Hans: 启用提示缓存
+      en_US: Enable Prompt Cache
+    type: boolean
+    required: false
+    default: true
+    help:
+      zh_Hans: 启用提示缓存可以提高性能并降低成本。Claude 3.7 Sonnet支持在system、messages和tools字段中使用缓存检查点。
+      en_US: Enable prompt caching to improve performance and reduce costs. Claude 3.7 Sonnet supports cache checkpoints in system, messages, and tools fields.
+  - name: reasoning_type
+    label:
+      zh_Hans: 推理配置
+      en_US: Reasoning Type
+    type: boolean
+    required: false
+    default: false
+    placeholder:
+      zh_Hans: 设置推理配置
+      en_US: Set reasoning configuration
+    help:
+      zh_Hans: 控制模型的推理能力。启用时，temperature将固定为1且top_p将被禁用。
+      en_US: Controls the model's reasoning capability. When enabled, temperature will be fixed to 1 and top_p will be disabled.
+  - name: reasoning_budget
+    show_on:
+      - variable: reasoning_type
+        value: true
+    label:
+      zh_Hans: 推理预算
+      en_US: Reasoning Budget
+    type: int
+    default: 1024
+    min: 0
+    max: 128000
+    help:
+      zh_Hans: 推理的预算限制（最小1024），必须小于max_tokens。仅在推理类型为enabled时可用。
+      en_US: Budget limit for reasoning (minimum 1024), must be less than max_tokens. Only available when reasoning type is enabled.
+
+  - name: max_tokens
+    use_template: max_tokens
+    required: true
+    label:
+      zh_Hans: 最大token数
+      en_US: Max Tokens
+    type: int
+    default: 8192
+    min: 1
+    max: 128000
+    help:
+      zh_Hans: 停止前生成的最大令牌数。请注意，Anthropic Claude 模型可能会在达到 max_tokens 的值之前停止生成令牌。不同的 Anthropic Claude 模型对此参数具有不同的最大值。
+      en_US: The maximum number of tokens to generate before stopping. Note that Anthropic Claude models might stop generating tokens before reaching the value of max_tokens. Different Anthropic Claude models have different maximum values for this parameter.
+  - name: temperature
+    use_template: temperature
+    required: false
+    label:
+      zh_Hans: 模型温度
+      en_US: Model Temperature
+    type: float
+    default: 1
+    min: 0.0
+    max: 1.0
+    help:
+      zh_Hans: 生成内容的随机性。当推理功能启用时，该值将被固定为1。
+      en_US: The amount of randomness injected into the response. When reasoning is enabled, this value will be fixed to 1.
+  - name: top_p
+    show_on:
+      - variable: reasoning_type
+        value: disabled
+    use_template: top_p
+    label:
+      zh_Hans: Top P
+      en_US: Top P
+    required: false
+    type: float
+    default: 0.999
+    min: 0.000
+    max: 1.000
+    help:
+      zh_Hans: 在核采样中的概率阈值。当推理功能启用时，该参数将被禁用。
+      en_US: The probability threshold in nucleus sampling. When reasoning is enabled, this parameter will be disabled.
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    required: false
+    type: int
+    default: 0
+    min: 0
+    # tip docs from aws has error, max value is 500
+    max: 500
+    help:
+      zh_Hans: 对于每个后续标记，仅从前 K 个选项中进行采样。使用 top_k 删除长尾低概率响应。
+      en_US: Only sample from the top K options for each subsequent token. Use top_k to remove long tail low probability responses.
+  - name: response_format
+    use_template: response_format
+pricing:
+  input: '0.003'
+  output: '0.015'
+  unit: '0.001'
+  currency: USD

api/core/model_runtime/model_providers/bedrock/llm/llm.py

@@ -58,6 +58,7 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
     # TODO There is invoke issue: context limit on Cohere Model, will add them after fixed.
     CONVERSE_API_ENABLED_MODEL_INFO = [
         {"prefix": "anthropic.claude-v2", "support_system_prompts": True, "support_tool_use": False},
+        {"prefix": "us.deepseek", "support_system_prompts": True, "support_tool_use": False},
         {"prefix": "anthropic.claude-v1", "support_system_prompts": True, "support_tool_use": False},
         {"prefix": "us.anthropic.claude-3", "support_system_prompts": True, "support_tool_use": True},
         {"prefix": "eu.anthropic.claude-3", "support_system_prompts": True, "support_tool_use": True},

api/core/model_runtime/model_providers/bedrock/llm/us.deepseek-r1.yaml

@@ -0,0 +1,63 @@
+model: us.deepseek.r1-v1:0
+label:
+  en_US: DeepSeek-R1(US.Cross Region Inference)
+icon: icon_s_en.svg
+model_type: llm
+features:
+  - agent-thought
+  - vision
+  - tool-call
+  - stream-tool-call
+model_properties:
+  mode: chat
+  context_size: 32768
+parameter_rules:
+  - name: max_tokens
+    use_template: max_tokens
+    required: true
+    label:
+      zh_Hans: 最大token数
+      en_US: Max Tokens
+    type: int
+    default: 8192
+    min: 1
+    max: 128000
+    help:
+      zh_Hans: 停止前生成的最大令牌数。
+      en_US: The maximum number of tokens to generate before stopping.
+  - name: temperature
+    use_template: temperature
+    required: false
+    label:
+      zh_Hans: 模型温度
+      en_US: Model Temperature
+    type: float
+    default: 1
+    min: 0.0
+    max: 1.0
+    help:
+      zh_Hans: 生成内容的随机性。当推理功能启用时，该值将被固定为1。
+      en_US: The amount of randomness injected into the response. When reasoning is enabled, this value will be fixed to 1.
+  - name: top_p
+    show_on:
+      - variable: reasoning_type
+        value: disabled
+    use_template: top_p
+    label:
+      zh_Hans: Top P
+      en_US: Top P
+    required: false
+    type: float
+    default: 0.999
+    min: 0.000
+    max: 1.000
+    help:
+      zh_Hans: 在核采样中的概率阈值。当推理功能启用时，该参数将被禁用。
+      en_US: The probability threshold in nucleus sampling. When reasoning is enabled, this parameter will be disabled.
+  - name: response_format
+    use_template: response_format
+pricing:
+  input: '0.001'
+  output: '0.005'
+  unit: '0.001'
+  currency: USD

api/core/model_runtime/model_providers/openai/llm/_position.yaml

@@ -1,3 +1,4 @@
+- gpt-4.1
 - o1
 - o1-2024-12-17
 - o1-mini

api/core/model_runtime/model_providers/openai/llm/gpt-4-1.yaml

@@ -0,0 +1,60 @@
+model: gpt-4.1
+label:
+  zh_Hans: gpt-4.1
+  en_US: gpt-4.1
+model_type: llm
+features:
+  - multi-tool-call
+  - agent-thought
+  - stream-tool-call
+  - vision
+model_properties:
+  mode: chat
+  context_size: 1047576
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: presence_penalty
+    use_template: presence_penalty
+  - name: frequency_penalty
+    use_template: frequency_penalty
+  - name: max_tokens
+    use_template: max_tokens
+    default: 512
+    min: 1
+    max: 32768
+  - name: reasoning_effort
+    label:
+      zh_Hans: 推理工作
+      en_US: Reasoning Effort
+    type: string
+    help:
+      zh_Hans: 限制推理模型的推理工作
+      en_US: Constrains effort on reasoning for reasoning models
+    required: false
+    options:
+      - low
+      - medium
+      - high
+  - name: response_format
+    label:
+      zh_Hans: 回复格式
+      en_US: Response Format
+    type: string
+    help:
+      zh_Hans: 指定模型必须输出的格式
+      en_US: specifying the format that the model must output
+    required: false
+    options:
+      - text
+      - json_object
+      - json_schema
+  - name: json_schema
+    use_template: json_schema
+pricing:
+  input: '2.00'
+  output: '8.00'
+  unit: '0.000001'
+  currency: USD

api/core/model_runtime/model_providers/openai/llm/llm.py

@@ -1057,7 +1057,7 @@ class OpenAILargeLanguageModel(_CommonOpenAI, LargeLanguageModel):
             model = "gpt-4o"
 
         try:
-            encoding = tiktoken.encoding_for_model(model)
+            encoding = tiktoken.get_encoding(model)
         except KeyError:
             logger.warning("Warning: model not found. Using cl100k_base encoding.")
             model = "cl100k_base"

api/core/workflow/nodes/code/code_node.py

@@ -195,7 +195,7 @@ class CodeNode(BaseNode[CodeNodeData]):
             if output_config.type == "object":
                 # check if output is object
                 if not isinstance(result.get(output_name), dict):
-                    if isinstance(result.get(output_name), type(None)):
+                    if result.get(output_name) is None:
                         transformed_result[output_name] = None
                     else:
                         raise OutputValidationError(
@@ -223,7 +223,7 @@ class CodeNode(BaseNode[CodeNodeData]):
             elif output_config.type == "array[number]":
                 # check if array of number available
                 if not isinstance(result[output_name], list):
-                    if isinstance(result[output_name], type(None)):
+                    if result[output_name] is None:
                         transformed_result[output_name] = None
                     else:
                         raise OutputValidationError(
@@ -244,7 +244,7 @@ class CodeNode(BaseNode[CodeNodeData]):
             elif output_config.type == "array[string]":
                 # check if array of string available
                 if not isinstance(result[output_name], list):
-                    if isinstance(result[output_name], type(None)):
+                    if result[output_name] is None:
                         transformed_result[output_name] = None
                     else:
                         raise OutputValidationError(
@@ -265,7 +265,7 @@ class CodeNode(BaseNode[CodeNodeData]):
             elif output_config.type == "array[object]":
                 # check if array of object available
                 if not isinstance(result[output_name], list):
-                    if isinstance(result[output_name], type(None)):
+                    if result[output_name] is None:
                         transformed_result[output_name] = None
                     else:
                         raise OutputValidationError(

api/core/workflow/nodes/llm/node.py

@@ -968,14 +968,12 @@ def _handle_memory_chat_mode(
     *,
     memory: TokenBufferMemory | None,
     memory_config: MemoryConfig | None,
-    model_config: ModelConfigWithCredentialsEntity,
+    model_config: ModelConfigWithCredentialsEntity,  # TODO(-LAN-): Needs to remove
 ) -> Sequence[PromptMessage]:
     memory_messages: Sequence[PromptMessage] = []
     # Get messages from memory for chat model
     if memory and memory_config:
-        rest_tokens = _calculate_rest_token(prompt_messages=[], model_config=model_config)
         memory_messages = memory.get_history_prompt_messages(
-            max_token_limit=rest_tokens,
             message_limit=memory_config.window.size if memory_config.window.enabled else None,
         )
     return memory_messages

api/poetry.lock

@@ -10473,44 +10473,44 @@ client = ["SQLAlchemy (>=1.4,<3)"]
 
 [[package]]
 name = "tiktoken"
-version = "0.8.0"
+version = "0.9.0"
 description = "tiktoken is a fast BPE tokeniser for use with OpenAI's models"
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 markers = "python_version == \"3.11\" or python_version >= \"3.12\""
 files = [
-    {file = "tiktoken-0.8.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:b07e33283463089c81ef1467180e3e00ab00d46c2c4bbcef0acab5f771d6695e"},
+    {file = "tiktoken-0.9.0-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:586c16358138b96ea804c034b8acf3f5d3f0258bd2bc3b0227af4af5d622e382"},
-    {file = "tiktoken-0.8.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9269348cb650726f44dd3bbb3f9110ac19a8dcc8f54949ad3ef652ca22a38e21"},
+    {file = "tiktoken-0.9.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:d9c59ccc528c6c5dd51820b3474402f69d9a9e1d656226848ad68a8d5b2e5108"},
-    {file = "tiktoken-0.8.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:25e13f37bc4ef2d012731e93e0fef21dc3b7aea5bb9009618de9a4026844e560"},
+    {file = "tiktoken-0.9.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f0968d5beeafbca2a72c595e8385a1a1f8af58feaebb02b227229b69ca5357fd"},
-    {file = "tiktoken-0.8.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f13d13c981511331eac0d01a59b5df7c0d4060a8be1e378672822213da51e0a2"},
+    {file = "tiktoken-0.9.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:92a5fb085a6a3b7350b8fc838baf493317ca0e17bd95e8642f95fc69ecfed1de"},
-    {file = "tiktoken-0.8.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:6b2ddbc79a22621ce8b1166afa9f9a888a664a579350dc7c09346a3b5de837d9"},
+    {file = "tiktoken-0.9.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:15a2752dea63d93b0332fb0ddb05dd909371ededa145fe6a3242f46724fa7990"},
-    {file = "tiktoken-0.8.0-cp310-cp310-win_amd64.whl", hash = "sha256:d8c2d0e5ba6453a290b86cd65fc51fedf247e1ba170191715b049dac1f628005"},
+    {file = "tiktoken-0.9.0-cp310-cp310-win_amd64.whl", hash = "sha256:26113fec3bd7a352e4b33dbaf1bd8948de2507e30bd95a44e2b1156647bc01b4"},
-    {file = "tiktoken-0.8.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:d622d8011e6d6f239297efa42a2657043aaed06c4f68833550cac9e9bc723ef1"},
+    {file = "tiktoken-0.9.0-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:f32cc56168eac4851109e9b5d327637f15fd662aa30dd79f964b7c39fbadd26e"},
-    {file = "tiktoken-0.8.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:2efaf6199717b4485031b4d6edb94075e4d79177a172f38dd934d911b588d54a"},
+    {file = "tiktoken-0.9.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:45556bc41241e5294063508caf901bf92ba52d8ef9222023f83d2483a3055348"},
-    {file = "tiktoken-0.8.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5637e425ce1fc49cf716d88df3092048359a4b3bbb7da762840426e937ada06d"},
+    {file = "tiktoken-0.9.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:03935988a91d6d3216e2ec7c645afbb3d870b37bcb67ada1943ec48678e7ee33"},
-    {file = "tiktoken-0.8.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9fb0e352d1dbe15aba082883058b3cce9e48d33101bdaac1eccf66424feb5b47"},
+    {file = "tiktoken-0.9.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8b3d80aad8d2c6b9238fc1a5524542087c52b860b10cbf952429ffb714bc1136"},
-    {file = "tiktoken-0.8.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:56edfefe896c8f10aba372ab5706b9e3558e78db39dd497c940b47bf228bc419"},
+    {file = "tiktoken-0.9.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:b2a21133be05dc116b1d0372af051cd2c6aa1d2188250c9b553f9fa49301b336"},
-    {file = "tiktoken-0.8.0-cp311-cp311-win_amd64.whl", hash = "sha256:326624128590def898775b722ccc327e90b073714227175ea8febbc920ac0a99"},
+    {file = "tiktoken-0.9.0-cp311-cp311-win_amd64.whl", hash = "sha256:11a20e67fdf58b0e2dea7b8654a288e481bb4fc0289d3ad21291f8d0849915fb"},
-    {file = "tiktoken-0.8.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:881839cfeae051b3628d9823b2e56b5cc93a9e2efb435f4cf15f17dc45f21586"},
+    {file = "tiktoken-0.9.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:e88f121c1c22b726649ce67c089b90ddda8b9662545a8aeb03cfef15967ddd03"},
-    {file = "tiktoken-0.8.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:fe9399bdc3f29d428f16a2f86c3c8ec20be3eac5f53693ce4980371c3245729b"},
+    {file = "tiktoken-0.9.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:a6600660f2f72369acb13a57fb3e212434ed38b045fd8cc6cdd74947b4b5d210"},
-    {file = "tiktoken-0.8.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9a58deb7075d5b69237a3ff4bb51a726670419db6ea62bdcd8bd80c78497d7ab"},
+    {file = "tiktoken-0.9.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:95e811743b5dfa74f4b227927ed86cbc57cad4df859cb3b643be797914e41794"},
-    {file = "tiktoken-0.8.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d2908c0d043a7d03ebd80347266b0e58440bdef5564f84f4d29fb235b5df3b04"},
+    {file = "tiktoken-0.9.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:99376e1370d59bcf6935c933cb9ba64adc29033b7e73f5f7569f3aad86552b22"},
-    {file = "tiktoken-0.8.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:294440d21a2a51e12d4238e68a5972095534fe9878be57d905c476017bff99fc"},
+    {file = "tiktoken-0.9.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:badb947c32739fb6ddde173e14885fb3de4d32ab9d8c591cbd013c22b4c31dd2"},
-    {file = "tiktoken-0.8.0-cp312-cp312-win_amd64.whl", hash = "sha256:d8f3192733ac4d77977432947d563d7e1b310b96497acd3c196c9bddb36ed9db"},
+    {file = "tiktoken-0.9.0-cp312-cp312-win_amd64.whl", hash = "sha256:5a62d7a25225bafed786a524c1b9f0910a1128f4232615bf3f8257a73aaa3b16"},
-    {file = "tiktoken-0.8.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:02be1666096aff7da6cbd7cdaa8e7917bfed3467cd64b38b1f112e96d3b06a24"},
+    {file = "tiktoken-0.9.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:2b0e8e05a26eda1249e824156d537015480af7ae222ccb798e5234ae0285dbdb"},
-    {file = "tiktoken-0.8.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:c94ff53c5c74b535b2cbf431d907fc13c678bbd009ee633a2aca269a04389f9a"},
+    {file = "tiktoken-0.9.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:27d457f096f87685195eea0165a1807fae87b97b2161fe8c9b1df5bd74ca6f63"},
-    {file = "tiktoken-0.8.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6b231f5e8982c245ee3065cd84a4712d64692348bc609d84467c57b4b72dcbc5"},
+    {file = "tiktoken-0.9.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2cf8ded49cddf825390e36dd1ad35cd49589e8161fdcb52aa25f0583e90a3e01"},
-    {file = "tiktoken-0.8.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4177faa809bd55f699e88c96d9bb4635d22e3f59d635ba6fd9ffedf7150b9953"},
+    {file = "tiktoken-0.9.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cc156cb314119a8bb9748257a2eaebd5cc0753b6cb491d26694ed42fc7cb3139"},
-    {file = "tiktoken-0.8.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:5376b6f8dc4753cd81ead935c5f518fa0fbe7e133d9e25f648d8c4dabdd4bad7"},
+    {file = "tiktoken-0.9.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:cd69372e8c9dd761f0ab873112aba55a0e3e506332dd9f7522ca466e817b1b7a"},
-    {file = "tiktoken-0.8.0-cp313-cp313-win_amd64.whl", hash = "sha256:18228d624807d66c87acd8f25fc135665617cab220671eb65b50f5d70fa51f69"},
+    {file = "tiktoken-0.9.0-cp313-cp313-win_amd64.whl", hash = "sha256:5ea0edb6f83dc56d794723286215918c1cde03712cbbafa0348b33448faf5b95"},
-    {file = "tiktoken-0.8.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7e17807445f0cf1f25771c9d86496bd8b5c376f7419912519699f3cc4dc5c12e"},
+    {file = "tiktoken-0.9.0-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:c6386ca815e7d96ef5b4ac61e0048cd32ca5a92d5781255e13b31381d28667dc"},
-    {file = "tiktoken-0.8.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:886f80bd339578bbdba6ed6d0567a0d5c6cfe198d9e587ba6c447654c65b8edc"},
+    {file = "tiktoken-0.9.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:75f6d5db5bc2c6274b674ceab1615c1778e6416b14705827d19b40e6355f03e0"},
-    {file = "tiktoken-0.8.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6adc8323016d7758d6de7313527f755b0fc6c72985b7d9291be5d96d73ecd1e1"},
+    {file = "tiktoken-0.9.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e15b16f61e6f4625a57a36496d28dd182a8a60ec20a534c5343ba3cafa156ac7"},
-    {file = "tiktoken-0.8.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b591fb2b30d6a72121a80be24ec7a0e9eb51c5500ddc7e4c2496516dd5e3816b"},
+    {file = "tiktoken-0.9.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3ebcec91babf21297022882344c3f7d9eed855931466c3311b1ad6b64befb3df"},
-    {file = "tiktoken-0.8.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:845287b9798e476b4d762c3ebda5102be87ca26e5d2c9854002825d60cdb815d"},
+    {file = "tiktoken-0.9.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:e5fd49e7799579240f03913447c0cdfa1129625ebd5ac440787afc4345990427"},
-    {file = "tiktoken-0.8.0-cp39-cp39-win_amd64.whl", hash = "sha256:1473cfe584252dc3fa62adceb5b1c763c1874e04511b197da4e6de51d6ce5a02"},
+    {file = "tiktoken-0.9.0-cp39-cp39-win_amd64.whl", hash = "sha256:26242ca9dc8b58e875ff4ca078b9a94d2f0813e6a535dcd2205df5d49d927cc7"},
-    {file = "tiktoken-0.8.0.tar.gz", hash = "sha256:9ccbb2740f24542534369c5635cfd9b2b3c2490754a78ac8831d99f89f94eeb2"},
+    {file = "tiktoken-0.9.0.tar.gz", hash = "sha256:d02a5ca6a938e0490e1ff957bc48c8b078c88cb83977be1625b1fd8aac792c5d"},
 ]
 
 [package.dependencies]
@@ -12389,4 +12389,4 @@ cffi = ["cffi (>=1.11)"]
 [metadata]
 lock-version = "2.1"
 python-versions = ">=3.11,<3.13"
-content-hash = "d197cdff507a70323c1d6aca11609188f54970f67715af744fe6def15b7776fd"
+content-hash = "0df8aef68385b6596306fd18af317a835023d648eb5028cd57ec463f176e4c0f"

api/pyproject.toml

@@ -85,7 +85,7 @@ sentry-sdk = { version = "~1.44.1", extras = ["flask"] }
 sqlalchemy = "~2.0.29"
 starlette = "0.41.0"
 tencentcloud-sdk-python-hunyuan = "~3.0.1294"
-tiktoken = "~0.8.0"
+tiktoken = "^0.9.0"
 tokenizers = "~0.15.0"
 transformers = "~4.35.0"
 unstructured = { version = "~0.16.1", extras = ["docx", "epub", "md", "msg", "ppt", "pptx"] }

api/services/app_dsl_service.py

@@ -55,13 +55,19 @@ def _check_version_compatibility(imported_version: str) -> ImportStatus:
     except version.InvalidVersion:
         return ImportStatus.FAILED
 
-    # Compare major version and minor version
+    # If imported version is newer than current, always return PENDING
-    if current_ver.major != imported_ver.major or current_ver.minor != imported_ver.minor:
+    if imported_ver > current_ver:
         return ImportStatus.PENDING
 
-    if current_ver.micro != imported_ver.micro:
+    # If imported version is older than current's major, return PENDING
+    if imported_ver.major < current_ver.major:
+        return ImportStatus.PENDING
+
+    # If imported version is older than current's minor, return COMPLETED_WITH_WARNINGS
+    if imported_ver.minor < current_ver.minor:
         return ImportStatus.COMPLETED_WITH_WARNINGS
 
+    # If imported version equals or is older than current's micro, return COMPLETED
     return ImportStatus.COMPLETED
 
 

docker/docker-compose-template.yaml

@@ -2,7 +2,7 @@ x-shared-env: &shared-api-worker-env
 services:
   # API service
   api:
-    image: langgenius/dify-api:0.15.6
+    image: langgenius/dify-api:0.15.7
     restart: always
     environment:
       # Use the shared environment variables.
@@ -25,7 +25,7 @@ services:
   # worker service
   # The Celery worker for processing the queue.
   worker:
-    image: langgenius/dify-api:0.15.6
+    image: langgenius/dify-api:0.15.7
     restart: always
     environment:
       # Use the shared environment variables.
@@ -47,7 +47,7 @@ services:
 
   # Frontend web application.
   web:
-    image: langgenius/dify-web:0.15.6
+    image: langgenius/dify-web:0.15.7
     restart: always
     environment:
       CONSOLE_API_URL: ${CONSOLE_API_URL:-}

docker/docker-compose.yaml

@@ -394,7 +394,7 @@ x-shared-env: &shared-api-worker-env
 services:
   # API service
   api:
-    image: langgenius/dify-api:0.15.6
+    image: langgenius/dify-api:0.15.7
     restart: always
     environment:
       # Use the shared environment variables.
@@ -417,7 +417,7 @@ services:
   # worker service
   # The Celery worker for processing the queue.
   worker:
-    image: langgenius/dify-api:0.15.6
+    image: langgenius/dify-api:0.15.7
     restart: always
     environment:
       # Use the shared environment variables.
@@ -439,7 +439,7 @@ services:
 
   # Frontend web application.
   web:
-    image: langgenius/dify-web:0.15.6
+    image: langgenius/dify-web:0.15.7
     restart: always
     environment:
       CONSOLE_API_URL: ${CONSOLE_API_URL:-}

web/package.json

@@ -1,6 +1,6 @@
 {
   "name": "dify-web",
-  "version": "0.15.6",
+  "version": "0.15.7",
   "private": true,
   "engines": {
     "node": ">=18.17.0"