chore: update version to 0.15.2 in packaging and docker configurations (#12940)

Signed-off-by: -LAN- <laipz8200@outlook.com>

.github/actions/setup-poetry/action.yml

@@ -8,7 +8,7 @@ inputs:
   poetry-version:
     description: Poetry version to set up
     required: true
-    default: '1.8.4'
+    default: '2.0.1'
   poetry-lockfile:
     description: Path to the Poetry lockfile to restore cache from
     required: true

.github/workflows/api-tests.yml

@@ -42,25 +42,23 @@ jobs:
         run: poetry install -C api --with dev
 
       - name: Check dependencies in pyproject.toml
-        run: poetry run -C api bash dev/pytest/pytest_artifacts.sh
+        run: poetry run -P api bash dev/pytest/pytest_artifacts.sh
 
       - name: Run Unit tests
-        run: poetry run -C api bash dev/pytest/pytest_unit_tests.sh
+        run: poetry run -P api bash dev/pytest/pytest_unit_tests.sh
 
       - name: Run ModelRuntime
-        run: poetry run -C api bash dev/pytest/pytest_model_runtime.sh
+        run: poetry run -P api bash dev/pytest/pytest_model_runtime.sh
 
       - name: Run dify config tests
-        run: poetry run -C api python dev/pytest/pytest_config_tests.py
+        run: poetry run -P api python dev/pytest/pytest_config_tests.py
 
       - name: Run Tool
-        run: poetry run -C api bash dev/pytest/pytest_tools.sh
+        run: poetry run -P api bash dev/pytest/pytest_tools.sh
 
       - name: Run mypy
         run: |
-          pushd api
-          poetry run python -m mypy --install-types --non-interactive .
-          popd
+          poetry run -C api python -m mypy --install-types --non-interactive .
 
       - name: Set up dotenvs
         run: |
@@ -80,4 +78,4 @@ jobs:
             ssrf_proxy
 
       - name: Run Workflow
-        run: poetry run -C api bash dev/pytest/pytest_workflow.sh
+        run: poetry run -P api bash dev/pytest/pytest_workflow.sh

.github/workflows/style.yml

@@ -38,12 +38,12 @@ jobs:
         if: steps.changed-files.outputs.any_changed == 'true'
         run: |
           poetry run -C api ruff --version
-          poetry run -C api ruff check ./api
-          poetry run -C api ruff format --check ./api
+          poetry run -C api ruff check ./
+          poetry run -C api ruff format --check ./
 
       - name: Dotenv check
         if: steps.changed-files.outputs.any_changed == 'true'
-        run: poetry run -C api dotenv-linter ./api/.env.example ./web/.env.example
+        run: poetry run -P api dotenv-linter ./api/.env.example ./web/.env.example
 
       - name: Lint hints
         if: failure()

.github/workflows/vdb-tests.yml

@@ -70,4 +70,4 @@ jobs:
             tidb
 
       - name: Test Vector Stores
-        run: poetry run -C api bash dev/pytest/pytest_vdb.sh
+        run: poetry run -P api bash dev/pytest/pytest_vdb.sh

api/.ruff.toml

@@ -53,10 +53,12 @@ ignore = [
     "FURB152", # math-constant
     "UP007", # non-pep604-annotation
     "UP032", # f-string
+    "UP045", # non-pep604-annotation-optional
     "B005", # strip-with-multi-characters
     "B006", # mutable-argument-default
     "B007", # unused-loop-control-variable
     "B026", # star-arg-unpacking-after-keyword-arg
+    "B903", # class-as-data-structure
     "B904", # raise-without-from-inside-except
     "B905", # zip-without-explicit-strict
     "N806", # non-lowercase-variable-in-function

api/Dockerfile

@@ -4,7 +4,7 @@ FROM python:3.12-slim-bookworm AS base
 WORKDIR /app/api
 
 # Install Poetry
-ENV POETRY_VERSION=1.8.4
+ENV POETRY_VERSION=2.0.1
 
 # if you located in China, you can use aliyun mirror to speed up
 # RUN pip install --no-cache-dir poetry==${POETRY_VERSION} -i https://mirrors.aliyun.com/pypi/simple/

api/README.md

@@ -79,5 +79,5 @@
 2. Run the tests locally with mocked system environment variables in `tool.pytest_env` section in `pyproject.toml`
 
    ```bash
-   poetry run -C api bash dev/pytest/pytest_all_tests.sh
+   poetry run -P api bash dev/pytest/pytest_all_tests.sh
    ```

api/configs/feature/__init__.py

@@ -146,7 +146,7 @@ class EndpointConfig(BaseSettings):
     )
 
     CONSOLE_WEB_URL: str = Field(
-        description="Base URL for the console web interface," "used for frontend references and CORS configuration",
+        description="Base URL for the console web interface,used for frontend references and CORS configuration",
         default="",
     )
 

api/configs/feature/hosted_service/__init__.py

@@ -181,7 +181,7 @@ class HostedFetchAppTemplateConfig(BaseSettings):
     """
 
     HOSTED_FETCH_APP_TEMPLATES_MODE: str = Field(
-        description="Mode for fetching app templates: remote, db, or builtin" " default to remote,",
+        description="Mode for fetching app templates: remote, db, or builtin default to remote,",
         default="remote",
     )
 

api/configs/packaging/__init__.py

@@ -9,7 +9,7 @@ class PackagingInfo(BaseSettings):
 
     CURRENT_VERSION: str = Field(
         description="Dify version",
-        default="0.15.1",
+        default="0.15.2",
     )
 
     COMMIT_SHA: str = Field(

api/controllers/console/admin.py

@@ -56,7 +56,7 @@ class InsertExploreAppListApi(Resource):
 
         app = App.query.filter(App.id == args["app_id"]).first()
         if not app:
-            raise NotFound(f'App \'{args["app_id"]}\' is not found')
+            raise NotFound(f"App '{args['app_id']}' is not found")
 
         site = app.site
         if not site:

api/controllers/console/datasets/datasets.py

@@ -457,7 +457,7 @@ class DatasetIndexingEstimateApi(Resource):
             )
         except LLMBadRequestError:
             raise ProviderNotInitializeError(
-                "No Embedding Model available. Please configure a valid provider " "in the Settings -> Model Provider."
+                "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
             )
         except ProviderTokenNotInitError as ex:
             raise ProviderNotInitializeError(ex.description)
@@ -619,8 +619,7 @@ class DatasetRetrievalSettingApi(Resource):
         vector_type = dify_config.VECTOR_STORE
         match vector_type:
             case (
-                VectorType.MILVUS
-                | VectorType.RELYT
+                VectorType.RELYT
                 | VectorType.PGVECTOR
                 | VectorType.TIDB_VECTOR
                 | VectorType.CHROMA
@@ -645,6 +644,7 @@ class DatasetRetrievalSettingApi(Resource):
                 | VectorType.TIDB_ON_QDRANT
                 | VectorType.LINDORM
                 | VectorType.COUCHBASE
+                | VectorType.MILVUS
             ):
                 return {
                     "retrieval_method": [

api/controllers/console/datasets/datasets_document.py

@@ -350,8 +350,7 @@ class DatasetInitApi(Resource):
                 )
             except InvokeAuthorizationError:
                 raise ProviderNotInitializeError(
-                    "No Embedding Model available. Please configure a valid provider "
-                    "in the Settings -> Model Provider."
+                    "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
                 )
             except ProviderTokenNotInitError as ex:
                 raise ProviderNotInitializeError(ex.description)
@@ -526,8 +525,7 @@ class DocumentBatchIndexingEstimateApi(DocumentResource):
                 return response.model_dump(), 200
             except LLMBadRequestError:
                 raise ProviderNotInitializeError(
-                    "No Embedding Model available. Please configure a valid provider "
-                    "in the Settings -> Model Provider."
+                    "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
                 )
             except ProviderTokenNotInitError as ex:
                 raise ProviderNotInitializeError(ex.description)

api/controllers/console/datasets/datasets_segments.py

@@ -168,8 +168,7 @@ class DatasetDocumentSegmentApi(Resource):
                 )
             except LLMBadRequestError:
                 raise ProviderNotInitializeError(
-                    "No Embedding Model available. Please configure a valid provider "
-                    "in the Settings -> Model Provider."
+                    "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
                 )
             except ProviderTokenNotInitError as ex:
                 raise ProviderNotInitializeError(ex.description)
@@ -217,8 +216,7 @@ class DatasetDocumentSegmentAddApi(Resource):
                 )
             except LLMBadRequestError:
                 raise ProviderNotInitializeError(
-                    "No Embedding Model available. Please configure a valid provider "
-                    "in the Settings -> Model Provider."
+                    "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
                 )
             except ProviderTokenNotInitError as ex:
                 raise ProviderNotInitializeError(ex.description)
@@ -267,8 +265,7 @@ class DatasetDocumentSegmentUpdateApi(Resource):
                 )
             except LLMBadRequestError:
                 raise ProviderNotInitializeError(
-                    "No Embedding Model available. Please configure a valid provider "
-                    "in the Settings -> Model Provider."
+                    "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
                 )
             except ProviderTokenNotInitError as ex:
                 raise ProviderNotInitializeError(ex.description)
@@ -437,8 +434,7 @@ class ChildChunkAddApi(Resource):
                 )
             except LLMBadRequestError:
                 raise ProviderNotInitializeError(
-                    "No Embedding Model available. Please configure a valid provider "
-                    "in the Settings -> Model Provider."
+                    "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
                 )
             except ProviderTokenNotInitError as ex:
                 raise ProviderNotInitializeError(ex.description)

api/controllers/service_api/dataset/segment.py

@@ -53,8 +53,7 @@ class SegmentApi(DatasetApiResource):
                 )
             except LLMBadRequestError:
                 raise ProviderNotInitializeError(
-                    "No Embedding Model available. Please configure a valid provider "
-                    "in the Settings -> Model Provider."
+                    "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
                 )
             except ProviderTokenNotInitError as ex:
                 raise ProviderNotInitializeError(ex.description)
@@ -95,8 +94,7 @@ class SegmentApi(DatasetApiResource):
                 )
             except LLMBadRequestError:
                 raise ProviderNotInitializeError(
-                    "No Embedding Model available. Please configure a valid provider "
-                    "in the Settings -> Model Provider."
+                    "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
                 )
             except ProviderTokenNotInitError as ex:
                 raise ProviderNotInitializeError(ex.description)
@@ -175,8 +173,7 @@ class DatasetSegmentApi(DatasetApiResource):
                 )
             except LLMBadRequestError:
                 raise ProviderNotInitializeError(
-                    "No Embedding Model available. Please configure a valid provider "
-                    "in the Settings -> Model Provider."
+                    "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
                 )
             except ProviderTokenNotInitError as ex:
                 raise ProviderNotInitializeError(ex.description)

api/controllers/service_api/wraps.py

@@ -195,7 +195,11 @@ def validate_and_get_api_token(scope: str | None = None):
     with Session(db.engine, expire_on_commit=False) as session:
         update_stmt = (
             update(ApiToken)
-            .where(ApiToken.token == auth_token, ApiToken.last_used_at < cutoff_time, ApiToken.type == scope)
+            .where(
+                ApiToken.token == auth_token,
+                (ApiToken.last_used_at.is_(None) | (ApiToken.last_used_at < cutoff_time)),
+                ApiToken.type == scope,
+            )
             .values(last_used_at=current_time)
             .returning(ApiToken)
         )

api/core/agent/cot_agent_runner.py

@@ -172,7 +172,7 @@ class CotAgentRunner(BaseAgentRunner, ABC):
 
             self.save_agent_thought(
                 agent_thought=agent_thought,
-                tool_name=scratchpad.action.action_name if scratchpad.action else "",
+                tool_name=(scratchpad.action.action_name if scratchpad.action and not scratchpad.is_final() else ""),
                 tool_input={scratchpad.action.action_name: scratchpad.action.action_input} if scratchpad.action else {},
                 tool_invoke_meta={},
                 thought=scratchpad.thought or "",

api/core/app/apps/base_app_queue_manager.py

@@ -167,8 +167,7 @@ class AppQueueManager:
         else:
             if isinstance(data, DeclarativeMeta) or hasattr(data, "_sa_instance_state"):
                 raise TypeError(
-                    "Critical Error: Passing SQLAlchemy Model instances "
-                    "that cause thread safety issues is not allowed."
+                    "Critical Error: Passing SQLAlchemy Model instances that cause thread safety issues is not allowed."
                 )
 
 

api/core/app/apps/message_based_app_generator.py

@@ -89,6 +89,7 @@ class MessageBasedAppGenerator(BaseAppGenerator):
             Conversation.id == conversation_id,
             Conversation.app_id == app_model.id,
             Conversation.status == "normal",
+            Conversation.is_deleted.is_(False),
         ]
 
         if isinstance(user, Account):

api/core/app/task_pipeline/message_cycle_manage.py

@@ -145,7 +145,7 @@ class MessageCycleManage:
 
             # get extension
             if "." in message_file.url:
-                extension = f'.{message_file.url.split(".")[-1]}'
+                extension = f".{message_file.url.split('.')[-1]}"
                 if len(extension) > 10:
                     extension = ".bin"
             else:

api/core/external_data_tool/api/api.py

@@ -62,8 +62,9 @@ class ApiExternalDataTool(ExternalDataTool):
 
         if not api_based_extension:
             raise ValueError(
-                "[External data tool] API query failed, variable: {}, "
-                "error: api_based_extension_id is invalid".format(self.variable)
+                "[External data tool] API query failed, variable: {}, error: api_based_extension_id is invalid".format(
+                    self.variable
+                )
             )
 
         # decrypt api_key

api/core/file/models.py

@@ -90,7 +90,7 @@ class File(BaseModel):
     def markdown(self) -> str:
         url = self.generate_url()
         if self.type == FileType.IMAGE:
-            text = f'![{self.filename or ""}]({url})'
+            text = f"![{self.filename or ''}]({url})"
         else:
             text = f"[{self.filename or url}]({url})"
 

api/core/llm_generator/prompts.py

@@ -131,7 +131,7 @@ JAVASCRIPT_CODE_GENERATOR_PROMPT_TEMPLATE = (
 SUGGESTED_QUESTIONS_AFTER_ANSWER_INSTRUCTION_PROMPT = (
     "Please help me predict the three most likely questions that human would ask, "
     "and keeping each question under 20 characters.\n"
-    "MAKE SURE your output is the SAME language as the Assistant's latest response"
+    "MAKE SURE your output is the SAME language as the Assistant's latest response. "
     "The output must be an array in JSON format following the specified schema:\n"
     '["question1","question2","question3"]\n'
 )

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

@@ -108,7 +108,7 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
         ai_model_entity = self._get_ai_model_entity(base_model_name=base_model_name, model=model)
 
         if not ai_model_entity:
-            raise CredentialsValidateFailedError(f'Base Model Name {credentials["base_model_name"]} is invalid')
+            raise CredentialsValidateFailedError(f"Base Model Name {credentials['base_model_name']} is invalid")
 
         try:
             client = AzureOpenAI(**self._to_credential_kwargs(credentials))

api/core/model_runtime/model_providers/azure_openai/text_embedding/text_embedding.py

@@ -130,7 +130,7 @@ class AzureOpenAITextEmbeddingModel(_CommonAzureOpenAI, TextEmbeddingModel):
             raise CredentialsValidateFailedError("Base Model Name is required")
 
         if not self._get_ai_model_entity(credentials["base_model_name"], model):
-            raise CredentialsValidateFailedError(f'Base Model Name {credentials["base_model_name"]} is invalid')
+            raise CredentialsValidateFailedError(f"Base Model Name {credentials['base_model_name']} is invalid")
 
         try:
             credentials_kwargs = self._to_credential_kwargs(credentials)

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

@@ -70,7 +70,7 @@ class BedrockRerankModel(RerankModel):
         rerankingConfiguration = {
             "type": "BEDROCK_RERANKING_MODEL",
             "bedrockRerankingConfiguration": {
-                "numberOfResults": top_n,
+                "numberOfResults": min(top_n, len(text_sources)),
                 "modelConfiguration": {
                     "modelArn": model_package_arn,
                 },

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

@@ -1,2 +1,3 @@
 - deepseek-chat
 - deepseek-coder
+- deepseek-reasoner

api/core/model_runtime/model_providers/deepseek/llm/deepseek-chat.yaml

@@ -10,7 +10,7 @@ features:
   - stream-tool-call
 model_properties:
   mode: chat
-  context_size: 128000
+  context_size: 64000
 parameter_rules:
   - name: temperature
     use_template: temperature

api/core/model_runtime/model_providers/deepseek/llm/deepseek-coder.yaml

@@ -10,7 +10,7 @@ features:
   - stream-tool-call
 model_properties:
   mode: chat
-  context_size: 128000
+  context_size: 64000
 parameter_rules:
   - name: temperature
     use_template: temperature

api/core/model_runtime/model_providers/deepseek/llm/deepseek-reasoner.yaml

@@ -0,0 +1,21 @@
+model: deepseek-reasoner
+label:
+  zh_Hans: deepseek-reasoner
+  en_US: deepseek-reasoner
+model_type: llm
+features:
+  - agent-thought
+model_properties:
+  mode: chat
+  context_size: 64000
+parameter_rules:
+  - name: max_tokens
+    use_template: max_tokens
+    min: 1
+    max: 8192
+    default: 4096
+pricing:
+  input: "4"
+  output: "16"
+  unit: "0.000001"
+  currency: RMB

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

@@ -24,9 +24,6 @@ class DeepseekLargeLanguageModel(OAIAPICompatLargeLanguageModel):
         user: Optional[str] = None,
     ) -> Union[LLMResult, Generator]:
         self._add_custom_parameters(credentials)
-        # {"response_format": "xx"} need convert to {"response_format": {"type": "xx"}}
-        if "response_format" in model_parameters:
-            model_parameters["response_format"] = {"type": model_parameters.get("response_format")}
         return super()._invoke(model, credentials, prompt_messages, model_parameters, tools, stop, stream)
 
     def validate_credentials(self, model: str, credentials: dict) -> None:

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

@@ -1,5 +1,6 @@
 - gemini-2.0-flash-exp
 - gemini-2.0-flash-thinking-exp-1219
+- gemini-2.0-flash-thinking-exp-01-21
 - gemini-1.5-pro
 - gemini-1.5-pro-latest
 - gemini-1.5-pro-001

api/core/model_runtime/model_providers/google/llm/gemini-2.0-flash-thinking-exp-01-21.yaml

@@ -0,0 +1,39 @@
+model: gemini-2.0-flash-thinking-exp-01-21
+label:
+  en_US: Gemini 2.0 Flash Thinking Exp 01-21
+model_type: llm
+features:
+  - agent-thought
+  - vision
+  - document
+  - video
+  - audio
+model_properties:
+  mode: chat
+  context_size: 32767
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: top_k
+    label:
+      zh_Hans: 取样数量
+      en_US: Top k
+    type: int
+    help:
+      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
+      en_US: Only sample from the top K options for each subsequent token.
+    required: false
+  - name: max_output_tokens
+    use_template: max_tokens
+    default: 8192
+    min: 1
+    max: 8192
+  - name: json_schema
+    use_template: json_schema
+pricing:
+  input: '0.00'
+  output: '0.00'
+  unit: '0.000001'
+  currency: USD

api/core/model_runtime/model_providers/huggingface_hub/text_embedding/text_embedding.py

@@ -162,9 +162,9 @@ class HuggingfaceHubTextEmbeddingModel(_CommonHuggingfaceHub, TextEmbeddingModel
     @staticmethod
     def _check_endpoint_url_model_repository_name(credentials: dict, model_name: str):
         try:
-            url = f'{HUGGINGFACE_ENDPOINT_API}{credentials["huggingface_namespace"]}'
+            url = f"{HUGGINGFACE_ENDPOINT_API}{credentials['huggingface_namespace']}"
             headers = {
-                "Authorization": f'Bearer {credentials["huggingfacehub_api_token"]}',
+                "Authorization": f"Bearer {credentials['huggingfacehub_api_token']}",
                 "Content-Type": "application/json",
             }
 

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

@@ -34,6 +34,7 @@ from core.model_runtime.model_providers.minimax.llm.types import MinimaxMessage
 
 class MinimaxLargeLanguageModel(LargeLanguageModel):
     model_apis = {
+        "minimax-text-01": MinimaxChatCompletionPro,
         "abab7-chat-preview": MinimaxChatCompletionPro,
         "abab6.5t-chat": MinimaxChatCompletionPro,
         "abab6.5s-chat": MinimaxChatCompletionPro,

api/core/model_runtime/model_providers/minimax/llm/minimax-text-01.yaml

@@ -0,0 +1,46 @@
+model: minimax-text-01
+label:
+  en_US: Minimax-Text-01
+model_type: llm
+features:
+  - agent-thought
+  - tool-call
+  - stream-tool-call
+model_properties:
+  mode: chat
+  context_size: 1000192
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+    min: 0.01
+    max: 1
+    default: 0.1
+  - name: top_p
+    use_template: top_p
+    min: 0.01
+    max: 1
+    default: 0.95
+  - name: max_tokens
+    use_template: max_tokens
+    required: true
+    default: 2048
+    min: 1
+    max: 1000192
+  - name: mask_sensitive_info
+    type: boolean
+    default: true
+    label:
+      zh_Hans: 隐私保护
+      en_US: Moderate
+    help:
+      zh_Hans: 对输出中易涉及隐私问题的文本信息进行打码，目前包括但不限于邮箱、域名、链接、证件号、家庭住址等，默认true，即开启打码
+      en_US: Mask the sensitive info of the generated content, such as email/domain/link/address/phone/id..
+  - name: presence_penalty
+    use_template: presence_penalty
+  - name: frequency_penalty
+    use_template: frequency_penalty
+pricing:
+  input: '0.001'
+  output: '0.008'
+  unit: '0.001'
+  currency: RMB

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

@@ -44,9 +44,6 @@ class MoonshotLargeLanguageModel(OAIAPICompatLargeLanguageModel):
         self._add_custom_parameters(credentials)
         self._add_function_call(model, credentials)
         user = user[:32] if user else None
-        # {"response_format": "json_object"} need convert to {"response_format": {"type": "json_object"}}
-        if "response_format" in model_parameters:
-            model_parameters["response_format"] = {"type": model_parameters.get("response_format")}
         return super()._invoke(model, credentials, prompt_messages, model_parameters, tools, stop, stream, user)
 
     def validate_credentials(self, model: str, credentials: dict) -> None:

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

@@ -1,5 +1,6 @@
 import json
 import logging
+import re
 from collections.abc import Generator
 from typing import Any, Optional, Union, cast
 
@@ -621,11 +622,19 @@ class OpenAILargeLanguageModel(_CommonOpenAI, LargeLanguageModel):
         prompt_messages = self._clear_illegal_prompt_messages(model, prompt_messages)
 
         # o1 compatibility
+        block_as_stream = False
         if model.startswith("o1"):
             if "max_tokens" in model_parameters:
                 model_parameters["max_completion_tokens"] = model_parameters["max_tokens"]
                 del model_parameters["max_tokens"]
 
+            if re.match(r"^o1(-\d{4}-\d{2}-\d{2})?$", model):
+                if stream:
+                    block_as_stream = True
+                    stream = False
+                    if "stream_options" in extra_model_kwargs:
+                        del extra_model_kwargs["stream_options"]
+
             if "stop" in extra_model_kwargs:
                 del extra_model_kwargs["stop"]
 
@@ -642,7 +651,45 @@ class OpenAILargeLanguageModel(_CommonOpenAI, LargeLanguageModel):
         if stream:
             return self._handle_chat_generate_stream_response(model, credentials, response, prompt_messages, tools)
 
-        return self._handle_chat_generate_response(model, credentials, response, prompt_messages, tools)
+        block_result = self._handle_chat_generate_response(model, credentials, response, prompt_messages, tools)
+
+        if block_as_stream:
+            return self._handle_chat_block_as_stream_response(block_result, prompt_messages, stop)
+
+        return block_result
+
+    def _handle_chat_block_as_stream_response(
+        self,
+        block_result: LLMResult,
+        prompt_messages: list[PromptMessage],
+        stop: Optional[list[str]] = None,
+    ) -> Generator[LLMResultChunk, None, None]:
+        """
+        Handle llm chat response
+        :param model: model name
+        :param credentials: credentials
+        :param response: response
+        :param prompt_messages: prompt messages
+        :param tools: tools for tool calling
+        :return: llm response chunk generator
+        """
+        text = block_result.message.content
+        text = cast(str, text)
+
+        if stop:
+            text = self.enforce_stop_tokens(text, stop)
+
+        yield LLMResultChunk(
+            model=block_result.model,
+            prompt_messages=prompt_messages,
+            system_fingerprint=block_result.system_fingerprint,
+            delta=LLMResultChunkDelta(
+                index=0,
+                message=block_result.message,
+                finish_reason="stop",
+                usage=block_result.usage,
+            ),
+        )
 
     def _handle_chat_generate_response(
         self,

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

@@ -29,9 +29,6 @@ class SiliconflowLargeLanguageModel(OAIAPICompatLargeLanguageModel):
         user: Optional[str] = None,
     ) -> Union[LLMResult, Generator]:
         self._add_custom_parameters(credentials)
-        # {"response_format": "json_object"} need convert to {"response_format": {"type": "json_object"}}
-        if "response_format" in model_parameters:
-            model_parameters["response_format"] = {"type": model_parameters.get("response_format")}
         return super()._invoke(model, credentials, prompt_messages, model_parameters, tools, stop, stream)
 
     def validate_credentials(self, model: str, credentials: dict) -> None:

api/core/model_runtime/model_providers/spark/llm/_client.py

@@ -21,7 +21,7 @@ class SparkLLMClient:
             domain = api_domain
 
         model_api_configs = {
-            "spark-lite": {"version": "v1.1", "chat_domain": "general"},
+            "spark-lite": {"version": "v1.1", "chat_domain": "lite"},
             "spark-pro": {"version": "v3.1", "chat_domain": "generalv3"},
             "spark-pro-128k": {"version": "pro-128k", "chat_domain": "pro-128k"},
             "spark-max": {"version": "v3.5", "chat_domain": "generalv3.5"},

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

@@ -257,8 +257,7 @@ class TongyiLargeLanguageModel(LargeLanguageModel):
         for index, response in enumerate(responses):
             if response.status_code not in {200, HTTPStatus.OK}:
                 raise ServiceUnavailableError(
-                    f"Failed to invoke model {model}, status code: {response.status_code}, "
-                    f"message: {response.message}"
+                    f"Failed to invoke model {model}, status code: {response.status_code}, message: {response.message}"
                 )
 
             resp_finish_reason = response.output.choices[0].finish_reason

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

@@ -146,7 +146,7 @@ class TritonInferenceAILargeLanguageModel(LargeLanguageModel):
             elif credentials["completion_type"] == "completion":
                 completion_type = LLMMode.COMPLETION.value
             else:
-                raise ValueError(f'completion_type {credentials["completion_type"]} is not supported')
+                raise ValueError(f"completion_type {credentials['completion_type']} is not supported")
 
         entity = AIModelEntity(
             model=model,

api/core/model_runtime/model_providers/volcengine_maas/llm/models.py

@@ -18,72 +18,93 @@ class ModelConfig(BaseModel):
 
 
 configs: dict[str, ModelConfig] = {
+    "Doubao-1.5-vision-pro-32k": ModelConfig(
+        properties=ModelProperties(context_size=32768, max_tokens=12288, mode=LLMMode.CHAT),
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.VISION],
+    ),
+    "Doubao-1.5-pro-32k": ModelConfig(
+        properties=ModelProperties(context_size=32768, max_tokens=12288, mode=LLMMode.CHAT),
+        features=[ModelFeature.AGENT_THOUGHT],
+    ),
+    "Doubao-1.5-lite-32k": ModelConfig(
+        properties=ModelProperties(context_size=32768, max_tokens=12288, mode=LLMMode.CHAT),
+        features=[ModelFeature.AGENT_THOUGHT],
+    ),
+    "Doubao-1.5-pro-256k": ModelConfig(
+        properties=ModelProperties(context_size=262144, max_tokens=12288, mode=LLMMode.CHAT),
+        features=[ModelFeature.AGENT_THOUGHT],
+    ),
     "Doubao-vision-pro-32k": ModelConfig(
         properties=ModelProperties(context_size=32768, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[ModelFeature.VISION],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.VISION],
     ),
     "Doubao-vision-lite-32k": ModelConfig(
         properties=ModelProperties(context_size=32768, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[ModelFeature.VISION],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.VISION],
     ),
     "Doubao-pro-4k": ModelConfig(
         properties=ModelProperties(context_size=4096, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[ModelFeature.TOOL_CALL],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.TOOL_CALL],
     ),
     "Doubao-lite-4k": ModelConfig(
         properties=ModelProperties(context_size=4096, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[ModelFeature.TOOL_CALL],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.TOOL_CALL],
     ),
     "Doubao-pro-32k": ModelConfig(
         properties=ModelProperties(context_size=32768, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[ModelFeature.TOOL_CALL],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.TOOL_CALL],
     ),
     "Doubao-lite-32k": ModelConfig(
         properties=ModelProperties(context_size=32768, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[ModelFeature.TOOL_CALL],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.TOOL_CALL],
     ),
     "Doubao-pro-256k": ModelConfig(
         properties=ModelProperties(context_size=262144, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[],
+        features=[ModelFeature.AGENT_THOUGHT],
     ),
     "Doubao-pro-128k": ModelConfig(
         properties=ModelProperties(context_size=131072, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[ModelFeature.TOOL_CALL],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.TOOL_CALL],
     ),
     "Doubao-lite-128k": ModelConfig(
-        properties=ModelProperties(context_size=131072, max_tokens=4096, mode=LLMMode.CHAT), features=[]
+        properties=ModelProperties(context_size=131072, max_tokens=4096, mode=LLMMode.CHAT),
+        features=[ModelFeature.AGENT_THOUGHT],
     ),
     "Skylark2-pro-4k": ModelConfig(
-        properties=ModelProperties(context_size=4096, max_tokens=4096, mode=LLMMode.CHAT), features=[]
+        properties=ModelProperties(context_size=4096, max_tokens=4096, mode=LLMMode.CHAT),
+        features=[ModelFeature.AGENT_THOUGHT],
     ),
     "Llama3-8B": ModelConfig(
-        properties=ModelProperties(context_size=8192, max_tokens=8192, mode=LLMMode.CHAT), features=[]
+        properties=ModelProperties(context_size=8192, max_tokens=8192, mode=LLMMode.CHAT),
+        features=[ModelFeature.AGENT_THOUGHT],
     ),
     "Llama3-70B": ModelConfig(
-        properties=ModelProperties(context_size=8192, max_tokens=8192, mode=LLMMode.CHAT), features=[]
+        properties=ModelProperties(context_size=8192, max_tokens=8192, mode=LLMMode.CHAT),
+        features=[ModelFeature.AGENT_THOUGHT],
     ),
     "Moonshot-v1-8k": ModelConfig(
         properties=ModelProperties(context_size=8192, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[ModelFeature.TOOL_CALL],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.TOOL_CALL],
     ),
     "Moonshot-v1-32k": ModelConfig(
         properties=ModelProperties(context_size=32768, max_tokens=16384, mode=LLMMode.CHAT),
-        features=[ModelFeature.TOOL_CALL],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.TOOL_CALL],
     ),
     "Moonshot-v1-128k": ModelConfig(
         properties=ModelProperties(context_size=131072, max_tokens=65536, mode=LLMMode.CHAT),
-        features=[ModelFeature.TOOL_CALL],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.TOOL_CALL],
     ),
     "GLM3-130B": ModelConfig(
         properties=ModelProperties(context_size=8192, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[ModelFeature.TOOL_CALL],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.TOOL_CALL],
     ),
     "GLM3-130B-Fin": ModelConfig(
         properties=ModelProperties(context_size=8192, max_tokens=4096, mode=LLMMode.CHAT),
-        features=[ModelFeature.TOOL_CALL],
+        features=[ModelFeature.AGENT_THOUGHT, ModelFeature.TOOL_CALL],
     ),
     "Mistral-7B": ModelConfig(
-        properties=ModelProperties(context_size=8192, max_tokens=2048, mode=LLMMode.CHAT), features=[]
+        properties=ModelProperties(context_size=8192, max_tokens=2048, mode=LLMMode.CHAT),
+        features=[ModelFeature.AGENT_THOUGHT],
     ),
 }
 

api/core/model_runtime/model_providers/volcengine_maas/volcengine_maas.yaml

@@ -118,6 +118,30 @@ model_credential_schema:
       type: select
       required: true
       options:
+        - label:
+            en_US: Doubao-1.5-vision-pro-32k
+          value: Doubao-1.5-vision-pro-32k
+          show_on:
+            - variable: __model_type
+              value: llm
+        - label:
+            en_US: Doubao-1.5-pro-32k
+          value: Doubao-1.5-pro-32k
+          show_on:
+            - variable: __model_type
+              value: llm
+        - label:
+            en_US: Doubao-1.5-lite-32k
+          value: Doubao-1.5-lite-32k
+          show_on:
+            - variable: __model_type
+              value: llm
+        - label:
+            en_US: Doubao-1.5-pro-256k
+          value: Doubao-1.5-pro-256k
+          show_on:
+            - variable: __model_type
+              value: llm
         - label:
             en_US: Doubao-vision-pro-32k
           value: Doubao-vision-pro-32k

api/core/model_runtime/model_providers/wenxin/_common.py

@@ -41,15 +41,15 @@ class BaiduAccessToken:
         resp = response.json()
         if "error" in resp:
             if resp["error"] == "invalid_client":
-                raise InvalidAPIKeyError(f'Invalid API key or secret key: {resp["error_description"]}')
+                raise InvalidAPIKeyError(f"Invalid API key or secret key: {resp['error_description']}")
             elif resp["error"] == "unknown_error":
-                raise InternalServerError(f'Internal server error: {resp["error_description"]}')
+                raise InternalServerError(f"Internal server error: {resp['error_description']}")
             elif resp["error"] == "invalid_request":
-                raise BadRequestError(f'Bad request: {resp["error_description"]}')
+                raise BadRequestError(f"Bad request: {resp['error_description']}")
             elif resp["error"] == "rate_limit_exceeded":
-                raise RateLimitReachedError(f'Rate limit reached: {resp["error_description"]}')
+                raise RateLimitReachedError(f"Rate limit reached: {resp['error_description']}")
             else:
-                raise Exception(f'Unknown error: {resp["error_description"]}')
+                raise Exception(f"Unknown error: {resp['error_description']}")
 
         return resp["access_token"]
 

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

@@ -406,7 +406,7 @@ class XinferenceAILargeLanguageModel(LargeLanguageModel):
             elif credentials["completion_type"] == "completion":
                 completion_type = LLMMode.COMPLETION.value
             else:
-                raise ValueError(f'completion_type {credentials["completion_type"]} is not supported')
+                raise ValueError(f"completion_type {credentials['completion_type']} is not supported")
         else:
             extra_args = XinferenceHelper.get_xinference_extra_parameter(
                 server_url=credentials["server_url"],
@@ -472,7 +472,7 @@ class XinferenceAILargeLanguageModel(LargeLanguageModel):
         api_key = credentials.get("api_key") or "abc"
 
         client = OpenAI(
-            base_url=f'{credentials["server_url"]}/v1',
+            base_url=f"{credentials['server_url']}/v1",
             api_key=api_key,
             max_retries=int(credentials.get("max_retries") or DEFAULT_MAX_RETRIES),
             timeout=int(credentials.get("invoke_timeout") or DEFAULT_INVOKE_TIMEOUT),

api/core/provider_manager.py

@@ -22,7 +22,12 @@ from core.helper import encrypter
 from core.helper.model_provider_cache import ProviderCredentialsCache, ProviderCredentialsCacheType
 from core.helper.position_helper import is_filtered
 from core.model_runtime.entities.model_entities import ModelType
-from core.model_runtime.entities.provider_entities import CredentialFormSchema, FormType, ProviderEntity
+from core.model_runtime.entities.provider_entities import (
+    ConfigurateMethod,
+    CredentialFormSchema,
+    FormType,
+    ProviderEntity,
+)
 from core.model_runtime.model_providers import model_provider_factory
 from extensions import ext_hosting_provider
 from extensions.ext_database import db
@@ -835,11 +840,18 @@ class ProviderManager:
         :return:
         """
         # Get provider model credential secret variables
-        model_credential_secret_variables = self._extract_secret_variables(
-            provider_entity.model_credential_schema.credential_form_schemas
-            if provider_entity.model_credential_schema
-            else []
-        )
+        if ConfigurateMethod.PREDEFINED_MODEL in provider_entity.configurate_methods:
+            model_credential_secret_variables = self._extract_secret_variables(
+                provider_entity.provider_credential_schema.credential_form_schemas
+                if provider_entity.provider_credential_schema
+                else []
+            )
+        else:
+            model_credential_secret_variables = self._extract_secret_variables(
+                provider_entity.model_credential_schema.credential_form_schemas
+                if provider_entity.model_credential_schema
+                else []
+            )
 
         model_settings: list[ModelSettings] = []
         if not provider_model_settings:

api/core/rag/extractor/firecrawl/firecrawl_app.py

@@ -31,7 +31,7 @@ class FirecrawlApp:
                     "markdown": data.get("markdown"),
                 }
             else:
-                raise Exception(f'Failed to scrape URL. Error: {response_data["error"]}')
+                raise Exception(f"Failed to scrape URL. Error: {response_data['error']}")
 
         elif response.status_code in {402, 409, 500}:
             error_message = response.json().get("error", "Unknown error occurred")

api/core/rag/extractor/notion_extractor.py

@@ -358,8 +358,7 @@ class NotionExtractor(BaseExtractor):
 
         if not data_source_binding:
             raise Exception(
-                f"No notion data source binding found for tenant {tenant_id} "
-                f"and notion workspace {notion_workspace_id}"
+                f"No notion data source binding found for tenant {tenant_id} and notion workspace {notion_workspace_id}"
             )
 
         return cast(str, data_source_binding.access_token)

api/core/tools/provider/builtin/aippt/tools/aippt.py

@@ -127,7 +127,7 @@ class AIPPTGenerateToolAdapter:
 
         response = response.json()
         if response.get("code") != 0:
-            raise Exception(f'Failed to create task: {response.get("msg")}')
+            raise Exception(f"Failed to create task: {response.get('msg')}")
 
         return response.get("data", {}).get("id")
 
@@ -222,7 +222,7 @@ class AIPPTGenerateToolAdapter:
         elif model == "wenxin":
             response = response.json()
             if response.get("code") != 0:
-                raise Exception(f'Failed to generate content: {response.get("msg")}')
+                raise Exception(f"Failed to generate content: {response.get('msg')}")
 
             return response.get("data", "")
 
@@ -254,7 +254,7 @@ class AIPPTGenerateToolAdapter:
 
         response = response.json()
         if response.get("code") != 0:
-            raise Exception(f'Failed to generate ppt: {response.get("msg")}')
+            raise Exception(f"Failed to generate ppt: {response.get('msg')}")
 
         id = response.get("data", {}).get("id")
         cover_url = response.get("data", {}).get("cover_url")
@@ -270,7 +270,7 @@ class AIPPTGenerateToolAdapter:
 
         response = response.json()
         if response.get("code") != 0:
-            raise Exception(f'Failed to generate ppt: {response.get("msg")}')
+            raise Exception(f"Failed to generate ppt: {response.get('msg')}")
 
         export_code = response.get("data")
         if not export_code:
@@ -290,7 +290,7 @@ class AIPPTGenerateToolAdapter:
 
             response = response.json()
             if response.get("code") != 0:
-                raise Exception(f'Failed to generate ppt: {response.get("msg")}')
+                raise Exception(f"Failed to generate ppt: {response.get('msg')}")
 
             if response.get("msg") == "导出中":
                 current_iteration += 1
@@ -343,7 +343,7 @@ class AIPPTGenerateToolAdapter:
             raise Exception(f"Failed to connect to aippt: {response.text}")
         response = response.json()
         if response.get("code") != 0:
-            raise Exception(f'Failed to connect to aippt: {response.get("msg")}')
+            raise Exception(f"Failed to connect to aippt: {response.get('msg')}")
 
         token = response.get("data", {}).get("token")
         expire = response.get("data", {}).get("time_expire")
@@ -379,7 +379,7 @@ class AIPPTGenerateToolAdapter:
                 if cls._style_cache[key]["expire"] < now:
                     del cls._style_cache[key]
 
-            key = f'{credentials["aippt_access_key"]}#@#{user_id}'
+            key = f"{credentials['aippt_access_key']}#@#{user_id}"
             if key in cls._style_cache:
                 return cls._style_cache[key]["colors"], cls._style_cache[key]["styles"]
 
@@ -396,11 +396,11 @@ class AIPPTGenerateToolAdapter:
         response = response.json()
 
         if response.get("code") != 0:
-            raise Exception(f'Failed to connect to aippt: {response.get("msg")}')
+            raise Exception(f"Failed to connect to aippt: {response.get('msg')}")
 
         colors = [
             {
-                "id": f'id-{item.get("id")}',
+                "id": f"id-{item.get('id')}",
                 "name": item.get("name"),
                 "en_name": item.get("en_name", item.get("name")),
             }
@@ -408,7 +408,7 @@ class AIPPTGenerateToolAdapter:
         ]
         styles = [
             {
-                "id": f'id-{item.get("id")}',
+                "id": f"id-{item.get('id')}",
                 "name": item.get("title"),
             }
             for item in response.get("data", {}).get("suit_style") or []
@@ -454,7 +454,7 @@ class AIPPTGenerateToolAdapter:
         response = response.json()
 
         if response.get("code") != 0:
-            raise Exception(f'Failed to connect to aippt: {response.get("msg")}')
+            raise Exception(f"Failed to connect to aippt: {response.get('msg')}")
 
         if len(response.get("data", {}).get("list") or []) > 0:
             return response.get("data", {}).get("list")[0].get("id")

api/core/tools/provider/builtin/aws/tools/nova_reel.py

@@ -229,8 +229,7 @@ class NovaReelTool(BuiltinTool):
 
         if async_mode:
             return self.create_text_message(
-                f"Video generation started.\nInvocation ARN: {invocation_arn}\n"
-                f"Video will be available at: {video_uri}"
+                f"Video generation started.\nInvocation ARN: {invocation_arn}\nVideo will be available at: {video_uri}"
             )
 
         return self._wait_for_completion(bedrock, s3_client, invocation_arn)

api/core/tools/provider/builtin/baidu_translate/tools/fieldtranslate.py

@@ -65,7 +65,7 @@ class BaiduFieldTranslateTool(BuiltinTool, BaiduTranslateToolBase):
             if "trans_result" in result:
                 result_text = result["trans_result"][0]["dst"]
             else:
-                result_text = f'{result["error_code"]}: {result["error_msg"]}'
+                result_text = f"{result['error_code']}: {result['error_msg']}"
 
             return self.create_text_message(str(result_text))
         except requests.RequestException as e:

api/core/tools/provider/builtin/baidu_translate/tools/language.py

@@ -52,7 +52,7 @@ class BaiduLanguageTool(BuiltinTool, BaiduTranslateToolBase):
 
             result_text = ""
             if result["error_code"] != 0:
-                result_text = f'{result["error_code"]}: {result["error_msg"]}'
+                result_text = f"{result['error_code']}: {result['error_msg']}"
             else:
                 result_text = result["data"]["src"]
                 result_text = self.mapping_result(description_language, result_text)

api/core/tools/provider/builtin/baidu_translate/tools/translate.py

@@ -58,7 +58,7 @@ class BaiduTranslateTool(BuiltinTool, BaiduTranslateToolBase):
             if "trans_result" in result:
                 result_text = result["trans_result"][0]["dst"]
             else:
-                result_text = f'{result["error_code"]}: {result["error_msg"]}'
+                result_text = f"{result['error_code']}: {result['error_msg']}"
 
             return self.create_text_message(str(result_text))
         except requests.RequestException as e:

api/core/tools/provider/builtin/bing/tools/bing_web_search.py

@@ -30,7 +30,7 @@ class BingSearchTool(BuiltinTool):
         headers = {"Ocp-Apim-Subscription-Key": subscription_key, "Accept-Language": accept_language}
 
         query = quote(query)
-        server_url = f'{server_url}?q={query}&mkt={market_code}&count={limit}&responseFilter={",".join(filters)}'
+        server_url = f"{server_url}?q={query}&mkt={market_code}&count={limit}&responseFilter={','.join(filters)}"
         response = get(server_url, headers=headers)
 
         if response.status_code != 200:
@@ -47,23 +47,23 @@ class BingSearchTool(BuiltinTool):
             results = []
             if search_results:
                 for result in search_results:
-                    url = f': {result["url"]}' if "url" in result else ""
-                    results.append(self.create_text_message(text=f'{result["name"]}{url}'))
+                    url = f": {result['url']}" if "url" in result else ""
+                    results.append(self.create_text_message(text=f"{result['name']}{url}"))
 
             if entities:
                 for entity in entities:
-                    url = f': {entity["url"]}' if "url" in entity else ""
-                    results.append(self.create_text_message(text=f'{entity.get("name", "")}{url}'))
+                    url = f": {entity['url']}" if "url" in entity else ""
+                    results.append(self.create_text_message(text=f"{entity.get('name', '')}{url}"))
 
             if news:
                 for news_item in news:
-                    url = f': {news_item["url"]}' if "url" in news_item else ""
-                    results.append(self.create_text_message(text=f'{news_item.get("name", "")}{url}'))
+                    url = f": {news_item['url']}" if "url" in news_item else ""
+                    results.append(self.create_text_message(text=f"{news_item.get('name', '')}{url}"))
 
             if related_searches:
                 for related in related_searches:
-                    url = f': {related["displayText"]}' if "displayText" in related else ""
-                    results.append(self.create_text_message(text=f'{related.get("displayText", "")}{url}'))
+                    url = f": {related['displayText']}" if "displayText" in related else ""
+                    results.append(self.create_text_message(text=f"{related.get('displayText', '')}{url}"))
 
             return results
         elif result_type == "json":
@@ -106,29 +106,29 @@ class BingSearchTool(BuiltinTool):
             text = ""
             if search_results:
                 for i, result in enumerate(search_results):
-                    text += f'{i + 1}: {result.get("name", "")} - {result.get("snippet", "")}\n'
+                    text += f"{i + 1}: {result.get('name', '')} - {result.get('snippet', '')}\n"
 
             if computation and "expression" in computation and "value" in computation:
                 text += "\nComputation:\n"
-                text += f'{computation["expression"]} = {computation["value"]}\n'
+                text += f"{computation['expression']} = {computation['value']}\n"
 
             if entities:
                 text += "\nEntities:\n"
                 for entity in entities:
-                    url = f'- {entity["url"]}' if "url" in entity else ""
-                    text += f'{entity.get("name", "")}{url}\n'
+                    url = f"- {entity['url']}" if "url" in entity else ""
+                    text += f"{entity.get('name', '')}{url}\n"
 
             if news:
                 text += "\nNews:\n"
                 for news_item in news:
-                    url = f'- {news_item["url"]}' if "url" in news_item else ""
-                    text += f'{news_item.get("name", "")}{url}\n'
+                    url = f"- {news_item['url']}" if "url" in news_item else ""
+                    text += f"{news_item.get('name', '')}{url}\n"
 
             if related_searches:
                 text += "\n\nRelated Searches:\n"
                 for related in related_searches:
-                    url = f'- {related["webSearchUrl"]}' if "webSearchUrl" in related else ""
-                    text += f'{related.get("displayText", "")}{url}\n'
+                    url = f"- {related['webSearchUrl']}" if "webSearchUrl" in related else ""
+                    text += f"{related.get('displayText', '')}{url}\n"
 
             return self.create_text_message(text=self.summary(user_id=user_id, content=text))
 

api/core/tools/provider/builtin/did/did_appx.py

@@ -83,5 +83,5 @@ class DIDApp:
             if status["status"] == "done":
                 return status
             elif status["status"] == "error" or status["status"] == "rejected":
-                raise HTTPError(f'Talks {id} failed: {status["status"]} {status.get("error", {}).get("description")}')
+                raise HTTPError(f"Talks {id} failed: {status['status']} {status.get('error', {}).get('description')}")
             time.sleep(poll_interval)

api/core/tools/provider/builtin/email/tools/send.py

@@ -20,33 +20,33 @@ class SendEmailToolParameters(BaseModel):
     encrypt_method: str
 
 
-def send_mail(parmas: SendEmailToolParameters):
+def send_mail(params: SendEmailToolParameters):
     timeout = 60
     msg = MIMEMultipart("alternative")
-    msg["From"] = parmas.email_account
-    msg["To"] = parmas.sender_to
-    msg["Subject"] = parmas.subject
-    msg.attach(MIMEText(parmas.email_content, "plain"))
-    msg.attach(MIMEText(parmas.email_content, "html"))
+    msg["From"] = params.email_account
+    msg["To"] = params.sender_to
+    msg["Subject"] = params.subject
+    msg.attach(MIMEText(params.email_content, "plain"))
+    msg.attach(MIMEText(params.email_content, "html"))
 
     ctx = ssl.create_default_context()
 
-    if parmas.encrypt_method.upper() == "SSL":
+    if params.encrypt_method.upper() == "SSL":
         try:
-            with smtplib.SMTP_SSL(parmas.smtp_server, parmas.smtp_port, context=ctx, timeout=timeout) as server:
-                server.login(parmas.email_account, parmas.email_password)
-                server.sendmail(parmas.email_account, parmas.sender_to, msg.as_string())
+            with smtplib.SMTP_SSL(params.smtp_server, params.smtp_port, context=ctx, timeout=timeout) as server:
+                server.login(params.email_account, params.email_password)
+                server.sendmail(params.email_account, params.sender_to, msg.as_string())
                 return True
         except Exception as e:
             logging.exception("send email failed")
             return False
     else:  # NONE or TLS
         try:
-            with smtplib.SMTP(parmas.smtp_server, parmas.smtp_port, timeout=timeout) as server:
-                if parmas.encrypt_method.upper() == "TLS":
+            with smtplib.SMTP(params.smtp_server, params.smtp_port, timeout=timeout) as server:
+                if params.encrypt_method.upper() == "TLS":
                     server.starttls(context=ctx)
-                server.login(parmas.email_account, parmas.email_password)
-                server.sendmail(parmas.email_account, parmas.sender_to, msg.as_string())
+                server.login(params.email_account, params.email_password)
+                server.sendmail(params.email_account, params.sender_to, msg.as_string())
                 return True
         except Exception as e:
             logging.exception("send email failed")

api/core/tools/provider/builtin/firecrawl/firecrawl_appx.py

@@ -74,7 +74,7 @@ class FirecrawlApp:
         if response is None:
             raise HTTPError("Failed to initiate crawl after multiple retries")
         elif response.get("success") == False:
-            raise HTTPError(f'Failed to crawl: {response.get("error")}')
+            raise HTTPError(f"Failed to crawl: {response.get('error')}")
         job_id: str = response["id"]
         if wait:
             return self._monitor_job_status(job_id=job_id, poll_interval=poll_interval)
@@ -100,7 +100,7 @@ class FirecrawlApp:
             if status["status"] == "completed":
                 return status
             elif status["status"] == "failed":
-                raise HTTPError(f'Job {job_id} failed: {status["error"]}')
+                raise HTTPError(f"Job {job_id} failed: {status['error']}")
             time.sleep(poll_interval)
 
 

api/core/tools/provider/builtin/gaode/tools/gaode_weather.py

@@ -37,8 +37,9 @@ class GaodeRepositoriesTool(BuiltinTool):
                     CityCode = City_data["districts"][0]["adcode"]
                     weatherInfo_response = s.request(
                         method="GET",
-                        url="{url}/weather/weatherInfo?city={citycode}&extensions=all&key={apikey}&output=json"
-                        "".format(url=api_domain, citycode=CityCode, apikey=self.runtime.credentials.get("api_key")),
+                        url="{url}/weather/weatherInfo?city={citycode}&extensions=all&key={apikey}&output=json".format(
+                            url=api_domain, citycode=CityCode, apikey=self.runtime.credentials.get("api_key")
+                        ),
                     )
                     weatherInfo_data = weatherInfo_response.json()
                     if weatherInfo_response.status_code == 200 and weatherInfo_data.get("info") == "OK":

api/core/tools/provider/builtin/gitlab/tools/gitlab_files.py

@@ -11,19 +11,21 @@ class GitlabFilesTool(BuiltinTool):
     def _invoke(
         self, user_id: str, tool_parameters: dict[str, Any]
     ) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
-        project = tool_parameters.get("project", "")
         repository = tool_parameters.get("repository", "")
+        project = tool_parameters.get("project", "")
         branch = tool_parameters.get("branch", "")
         path = tool_parameters.get("path", "")
+        file_path = tool_parameters.get("file_path", "")
 
-        if not project and not repository:
-            return self.create_text_message("Either project or repository is required")
+        if not repository and not project:
+            return self.create_text_message("Either repository or project is required")
         if not branch:
             return self.create_text_message("Branch is required")
-        if not path:
-            return self.create_text_message("Path is required")
+        if not path and not file_path:
+            return self.create_text_message("Either path or file_path is required")
 
         access_token = self.runtime.credentials.get("access_tokens")
+        headers = {"PRIVATE-TOKEN": access_token}
         site_url = self.runtime.credentials.get("site_url")
 
         if "access_tokens" not in self.runtime.credentials or not self.runtime.credentials.get("access_tokens"):
@@ -31,33 +33,45 @@ class GitlabFilesTool(BuiltinTool):
         if "site_url" not in self.runtime.credentials or not self.runtime.credentials.get("site_url"):
             site_url = "https://gitlab.com"
 
-        # Get file content
         if repository:
-            result = self.fetch_files(site_url, access_token, repository, branch, path, is_repository=True)
+            # URL encode the repository path
+            identifier = urllib.parse.quote(repository, safe="")
         else:
-            result = self.fetch_files(site_url, access_token, project, branch, path, is_repository=False)
+            identifier = self.get_project_id(site_url, access_token, project)
+            if not identifier:
+                raise Exception(f"Project '{project}' not found.)")
 
-        return [self.create_json_message(item) for item in result]
+        # Get file content
+        if path:
+            results = self.fetch_files(site_url, headers, identifier, branch, path)
+            return [self.create_json_message(item) for item in results]
+        else:
+            result = self.fetch_file(site_url, headers, identifier, branch, file_path)
+            return [self.create_json_message(result)]
+
+    @staticmethod
+    def fetch_file(
+        site_url: str,
+        headers: dict[str, str],
+        identifier: str,
+        branch: str,
+        path: str,
+    ) -> dict[str, Any]:
+        encoded_file_path = urllib.parse.quote(path, safe="")
+        file_url = f"{site_url}/api/v4/projects/{identifier}/repository/files/{encoded_file_path}/raw?ref={branch}"
+
+        file_response = requests.get(file_url, headers=headers)
+        file_response.raise_for_status()
+        file_content = file_response.text
+        return {"path": path, "branch": branch, "content": file_content}
 
     def fetch_files(
-        self, site_url: str, access_token: str, identifier: str, branch: str, path: str, is_repository: bool
+        self, site_url: str, headers: dict[str, str], identifier: str, branch: str, path: str
     ) -> list[dict[str, Any]]:
-        domain = site_url
-        headers = {"PRIVATE-TOKEN": access_token}
         results = []
 
         try:
-            if is_repository:
-                # URL encode the repository path
-                encoded_identifier = urllib.parse.quote(identifier, safe="")
-                tree_url = f"{domain}/api/v4/projects/{encoded_identifier}/repository/tree?path={path}&ref={branch}"
-            else:
-                # Get project ID from project name
-                project_id = self.get_project_id(site_url, access_token, identifier)
-                if not project_id:
-                    return self.create_text_message(f"Project '{identifier}' not found.")
-                tree_url = f"{domain}/api/v4/projects/{project_id}/repository/tree?path={path}&ref={branch}"
-
+            tree_url = f"{site_url}/api/v4/projects/{identifier}/repository/tree?path={path}&ref={branch}"
             response = requests.get(tree_url, headers=headers)
             response.raise_for_status()
             items = response.json()
@@ -65,26 +79,10 @@ class GitlabFilesTool(BuiltinTool):
             for item in items:
                 item_path = item["path"]
                 if item["type"] == "tree":  # It's a directory
-                    results.extend(
-                        self.fetch_files(site_url, access_token, identifier, branch, item_path, is_repository)
-                    )
+                    results.extend(self.fetch_files(site_url, headers, identifier, branch, item_path))
                 else:  # It's a file
-                    encoded_item_path = urllib.parse.quote(item_path, safe="")
-                    if is_repository:
-                        file_url = (
-                            f"{domain}/api/v4/projects/{encoded_identifier}/repository/files"
-                            f"/{encoded_item_path}/raw?ref={branch}"
-                        )
-                    else:
-                        file_url = (
-                            f"{domain}/api/v4/projects/{project_id}/repository/files"
-                            f"{encoded_item_path}/raw?ref={branch}"
-                        )
-
-                    file_response = requests.get(file_url, headers=headers)
-                    file_response.raise_for_status()
-                    file_content = file_response.text
-                    results.append({"path": item_path, "branch": branch, "content": file_content})
+                    result = self.fetch_file(site_url, headers, identifier, branch, item_path)
+                    results.append(result)
         except requests.RequestException as e:
             print(f"Error fetching data from GitLab: {e}")
 

api/core/tools/provider/builtin/gitlab/tools/gitlab_files.yaml

@@ -29,7 +29,7 @@ parameters:
       zh_Hans: 项目
     human_description:
       en_US: project
-      zh_Hans: 项目
+      zh_Hans: 项目(和仓库路径二选一,都填写以仓库路径优先)
     llm_description: Project for GitLab
     form: llm
   - name: branch
@@ -45,12 +45,21 @@ parameters:
     form: llm
   - name: path
     type: string
-    required: true
     label:
       en_US: path
-      zh_Hans: 文件路径
+      zh_Hans: 文件夹
     human_description:
       en_US: path
+      zh_Hans: 文件夹
+    llm_description: Dir path for GitLab
+    form: llm
+  - name: file_path
+    type: string
+    label:
+      en_US: file_path
       zh_Hans: 文件路径
+    human_description:
+      en_US: file_path
+      zh_Hans: 文件路径(和文件夹二选一,都填写以文件夹优先)
     llm_description: File path for GitLab
     form: llm

api/core/tools/provider/builtin/hap/tools/list_worksheet_records.py

@@ -110,7 +110,7 @@ class ListWorksheetRecordsTool(BuiltinTool):
                             result["rows"].append(self.get_row_field_value(row, schema))
                         return self.create_text_message(json.dumps(result, ensure_ascii=False))
                     else:
-                        result_text = f"Found {result['total']} rows in worksheet \"{worksheet_name}\"."
+                        result_text = f'Found {result["total"]} rows in worksheet "{worksheet_name}".'
                         if result["total"] > 0:
                             result_text += (
                                 f" The following are {min(limit, result['total'])}"

api/core/tools/provider/builtin/stability/tools/base.py

@@ -28,4 +28,4 @@ class BaseStabilityAuthorization:
         """
         This method is responsible for generating the authorization headers.
         """
-        return {"Authorization": f'Bearer {credentials.get("api_key", "")}'}
+        return {"Authorization": f"Bearer {credentials.get('api_key', '')}"}

api/core/tools/provider/builtin/vanna/vanna.py

@@ -38,7 +38,7 @@ class VannaProvider(BuiltinToolProviderController):
                 tool_parameters={
                     "model": "chinook",
                     "db_type": "SQLite",
-                    "url": f'{self._get_protocol_and_main_domain(credentials["base_url"])}/Chinook.sqlite',
+                    "url": f"{self._get_protocol_and_main_domain(credentials['base_url'])}/Chinook.sqlite",
                     "query": "What are the top 10 customers by sales?",
                 },
             )

api/core/tools/provider/builtin/websearch/tools/job_search.py

@@ -43,7 +43,7 @@ class SerplyApi:
     def parse_results(res: dict) -> str:
         """Process response from Serply Job Search."""
         jobs = res.get("jobs", [])
-        if not jobs:
+        if not res or "jobs" not in res:
             raise ValueError(f"Got error from Serply: {res}")
 
         string = []

api/core/tools/provider/builtin/websearch/tools/news_search.py

@@ -43,7 +43,7 @@ class SerplyApi:
     def parse_results(res: dict) -> str:
         """Process response from Serply News Search."""
         news = res.get("entries", [])
-        if not news:
+        if not res or "entries" not in res:
             raise ValueError(f"Got error from Serply: {res}")
 
         string = []

api/core/tools/provider/builtin/websearch/tools/scholar_search.py

@@ -43,7 +43,7 @@ class SerplyApi:
     def parse_results(res: dict) -> str:
         """Process response from Serply News Search."""
         articles = res.get("articles", [])
-        if not articles:
+        if not res or "articles" not in res:
             raise ValueError(f"Got error from Serply: {res}")
 
         string = []

api/core/tools/provider/builtin/websearch/tools/web_search.py

@@ -42,7 +42,7 @@ class SerplyApi:
     def parse_results(res: dict) -> str:
         """Process response from Serply Web Search."""
         results = res.get("results", [])
-        if not results:
+        if not res or "results" not in res:
             raise ValueError(f"Got error from Serply: {res}")
 
         string = []

api/core/tools/tool/api_tool.py

@@ -84,9 +84,9 @@ class ApiTool(Tool):
             if "api_key_header_prefix" in credentials:
                 api_key_header_prefix = credentials["api_key_header_prefix"]
                 if api_key_header_prefix == "basic" and credentials["api_key_value"]:
-                    credentials["api_key_value"] = f'Basic {credentials["api_key_value"]}'
+                    credentials["api_key_value"] = f"Basic {credentials['api_key_value']}"
                 elif api_key_header_prefix == "bearer" and credentials["api_key_value"]:
-                    credentials["api_key_value"] = f'Bearer {credentials["api_key_value"]}'
+                    credentials["api_key_value"] = f"Bearer {credentials['api_key_value']}"
                 elif api_key_header_prefix == "custom":
                     pass
 

api/core/tools/utils/message_transformer.py

@@ -29,7 +29,7 @@ class ToolFileMessageTransformer:
                         user_id=user_id, tenant_id=tenant_id, conversation_id=conversation_id, file_url=message.message
                     )
 
-                    url = f'/files/tools/{file.id}{guess_extension(file.mimetype) or ".png"}'
+                    url = f"/files/tools/{file.id}{guess_extension(file.mimetype) or '.png'}"
 
                     result.append(
                         ToolInvokeMessage(
@@ -122,4 +122,4 @@ class ToolFileMessageTransformer:
 
     @classmethod
     def get_tool_file_url(cls, tool_file_id: str, extension: Optional[str]) -> str:
-        return f'/files/tools/{tool_file_id}{extension or ".bin"}'
+        return f"/files/tools/{tool_file_id}{extension or '.bin'}"

api/core/tools/utils/parser.py

@@ -149,7 +149,7 @@ class ApiBasedToolSchemaParser:
                 if not path:
                     path = str(uuid.uuid4())
 
-                interface["operation"]["operationId"] = f'{path}_{interface["method"]}'
+                interface["operation"]["operationId"] = f"{path}_{interface['method']}"
 
             bundles.append(
                 ApiToolBundle(

api/core/variables/segments.py

@@ -134,6 +134,10 @@ class ArrayStringSegment(ArraySegment):
     value_type: SegmentType = SegmentType.ARRAY_STRING
     value: Sequence[str]
 
+    @property
+    def text(self) -> str:
+        return json.dumps(self.value)
+
 
 class ArrayNumberSegment(ArraySegment):
     value_type: SegmentType = SegmentType.ARRAY_NUMBER

api/core/workflow/nodes/http_request/executor.py

@@ -253,9 +253,9 @@ class Executor:
         )
         if executor_response.size > threshold_size:
             raise ResponseSizeError(
-                f'{"File" if executor_response.is_file else "Text"} size is too large,'
-                f' max size is {threshold_size / 1024 / 1024:.2f} MB,'
-                f' but current size is {executor_response.readable_size}.'
+                f"{'File' if executor_response.is_file else 'Text'} size is too large,"
+                f" max size is {threshold_size / 1024 / 1024:.2f} MB,"
+                f" but current size is {executor_response.readable_size}."
             )
 
         return executor_response
@@ -338,7 +338,7 @@ class Executor:
                 if self.auth.config and self.auth.config.header:
                     authorization_header = self.auth.config.header
                 if k.lower() == authorization_header.lower():
-                    raw += f'{k}: {"*" * len(v)}\r\n'
+                    raw += f"{k}: {'*' * len(v)}\r\n"
                     continue
             raw += f"{k}: {v}\r\n"
 

api/events/event_handlers/delete_tool_parameters_cache_when_sync_draft_workflow.py

@@ -26,7 +26,7 @@ def handle(sender, **kwargs):
                     tool_runtime=tool_runtime,
                     provider_name=tool_entity.provider_name,
                     provider_type=tool_entity.provider_type,
-                    identity_id=f'WORKFLOW.{app.id}.{node_data.get("id")}',
+                    identity_id=f"WORKFLOW.{app.id}.{node_data.get('id')}",
                 )
                 manager.delete_tool_parameters_cache()
             except:

api/fields/member_fields.py

@@ -1,6 +1,6 @@
 from flask_restful import fields  # type: ignore
 
-from libs.helper import TimestampField
+from libs.helper import AvatarUrlField, TimestampField
 
 simple_account_fields = {"id": fields.String, "name": fields.String, "email": fields.String}
 
@@ -8,6 +8,7 @@ account_fields = {
     "id": fields.String,
     "name": fields.String,
     "avatar": fields.String,
+    "avatar_url": AvatarUrlField,
     "email": fields.String,
     "is_password_set": fields.Boolean,
     "interface_language": fields.String,
@@ -22,6 +23,7 @@ account_with_role_fields = {
     "id": fields.String,
     "name": fields.String,
     "avatar": fields.String,
+    "avatar_url": AvatarUrlField,
     "email": fields.String,
     "last_login_at": TimestampField,
     "last_active_at": TimestampField,

api/libs/helper.py

@@ -41,6 +41,18 @@ class AppIconUrlField(fields.Raw):
         return None
 
 
+class AvatarUrlField(fields.Raw):
+    def output(self, key, obj):
+        if obj is None:
+            return None
+
+        from models.account import Account
+
+        if isinstance(obj, Account) and obj.avatar is not None:
+            return file_helpers.get_signed_file_url(obj.avatar)
+        return None
+
+
 class TimestampField(fields.Raw):
     def format(self, value) -> int:
         return int(value.timestamp())

api/models/dataset.py

@@ -13,6 +13,7 @@ from typing import Any, cast
 
 from sqlalchemy import func
 from sqlalchemy.dialects.postgresql import JSONB
+from sqlalchemy.orm import Mapped
 
 from configs import dify_config
 from core.rag.retrieval.retrieval_methods import RetrievalMethod
@@ -515,7 +516,7 @@ class DocumentSegment(db.Model):  # type: ignore[name-defined]
     tenant_id = db.Column(StringUUID, nullable=False)
     dataset_id = db.Column(StringUUID, nullable=False)
     document_id = db.Column(StringUUID, nullable=False)
-    position = db.Column(db.Integer, nullable=False)
+    position: Mapped[int]
     content = db.Column(db.Text, nullable=False)
     answer = db.Column(db.Text, nullable=True)
     word_count = db.Column(db.Integer, nullable=False)

api/pyproject.toml

@@ -1,9 +1,10 @@
 [project]
 name = "dify-api"
 requires-python = ">=3.11,<3.13"
+dynamic = [ "dependencies" ]
 
 [build-system]
-requires = ["poetry-core"]
+requires = ["poetry-core>=2.0.0"]
 build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
@@ -191,4 +192,4 @@ pytest-mock = "~3.14.0"
 optional = true
 [tool.poetry.group.lint.dependencies]
 dotenv-linter = "~0.5.0"
-ruff = "~0.8.1"
+ruff = "~0.9.2"

api/services/dataset_service.py

@@ -4,6 +4,7 @@ import logging
 import random
 import time
 import uuid
+from collections import Counter
 from typing import Any, Optional
 
 from flask_login import current_user  # type: ignore
@@ -221,8 +222,7 @@ class DatasetService:
                 )
             except LLMBadRequestError:
                 raise ValueError(
-                    "No Embedding Model available. Please configure a valid provider "
-                    "in the Settings -> Model Provider."
+                    "No Embedding Model available. Please configure a valid provider in the Settings -> Model Provider."
                 )
             except ProviderTokenNotInitError as ex:
                 raise ValueError(f"The dataset in unavailable, due to: {ex.description}")
@@ -859,7 +859,7 @@ class DocumentService:
                 position = DocumentService.get_documents_position(dataset.id)
                 document_ids = []
                 duplicate_document_ids = []
-                if knowledge_config.data_source.info_list.data_source_type == "upload_file":
+                if knowledge_config.data_source.info_list.data_source_type == "upload_file":  # type: ignore
                     upload_file_list = knowledge_config.data_source.info_list.file_info_list.file_ids  # type: ignore
                     for file_id in upload_file_list:
                         file = (
@@ -901,7 +901,7 @@ class DocumentService:
                         document = DocumentService.build_document(
                             dataset,
                             dataset_process_rule.id,  # type: ignore
-                            knowledge_config.data_source.info_list.data_source_type,
+                            knowledge_config.data_source.info_list.data_source_type,  # type: ignore
                             knowledge_config.doc_form,
                             knowledge_config.doc_language,
                             data_source_info,
@@ -916,8 +916,8 @@ class DocumentService:
                         document_ids.append(document.id)
                         documents.append(document)
                         position += 1
-                elif knowledge_config.data_source.info_list.data_source_type == "notion_import":
-                    notion_info_list = knowledge_config.data_source.info_list.notion_info_list
+                elif knowledge_config.data_source.info_list.data_source_type == "notion_import":  # type: ignore
+                    notion_info_list = knowledge_config.data_source.info_list.notion_info_list  # type: ignore
                     if not notion_info_list:
                         raise ValueError("No notion info list found.")
                     exist_page_ids = []
@@ -956,7 +956,7 @@ class DocumentService:
                                 document = DocumentService.build_document(
                                     dataset,
                                     dataset_process_rule.id,  # type: ignore
-                                    knowledge_config.data_source.info_list.data_source_type,
+                                    knowledge_config.data_source.info_list.data_source_type,  # type: ignore
                                     knowledge_config.doc_form,
                                     knowledge_config.doc_language,
                                     data_source_info,
@@ -976,8 +976,8 @@ class DocumentService:
                     # delete not selected documents
                     if len(exist_document) > 0:
                         clean_notion_document_task.delay(list(exist_document.values()), dataset.id)
-                elif knowledge_config.data_source.info_list.data_source_type == "website_crawl":
-                    website_info = knowledge_config.data_source.info_list.website_info_list
+                elif knowledge_config.data_source.info_list.data_source_type == "website_crawl":  # type: ignore
+                    website_info = knowledge_config.data_source.info_list.website_info_list  # type: ignore
                     if not website_info:
                         raise ValueError("No website info list found.")
                     urls = website_info.urls
@@ -996,7 +996,7 @@ class DocumentService:
                         document = DocumentService.build_document(
                             dataset,
                             dataset_process_rule.id,  # type: ignore
-                            knowledge_config.data_source.info_list.data_source_type,
+                            knowledge_config.data_source.info_list.data_source_type,  # type: ignore
                             knowledge_config.doc_form,
                             knowledge_config.doc_language,
                             data_source_info,
@@ -1195,20 +1195,20 @@ class DocumentService:
 
         if features.billing.enabled:
             count = 0
-            if knowledge_config.data_source.info_list.data_source_type == "upload_file":
+            if knowledge_config.data_source.info_list.data_source_type == "upload_file":  # type: ignore
                 upload_file_list = (
-                    knowledge_config.data_source.info_list.file_info_list.file_ids
-                    if knowledge_config.data_source.info_list.file_info_list
+                    knowledge_config.data_source.info_list.file_info_list.file_ids  # type: ignore
+                    if knowledge_config.data_source.info_list.file_info_list  # type: ignore
                     else []
                 )
                 count = len(upload_file_list)
-            elif knowledge_config.data_source.info_list.data_source_type == "notion_import":
-                notion_info_list = knowledge_config.data_source.info_list.notion_info_list
+            elif knowledge_config.data_source.info_list.data_source_type == "notion_import":  # type: ignore
+                notion_info_list = knowledge_config.data_source.info_list.notion_info_list  # type: ignore
                 if notion_info_list:
                     for notion_info in notion_info_list:
                         count = count + len(notion_info.pages)
-            elif knowledge_config.data_source.info_list.data_source_type == "website_crawl":
-                website_info = knowledge_config.data_source.info_list.website_info_list
+            elif knowledge_config.data_source.info_list.data_source_type == "website_crawl":  # type: ignore
+                website_info = knowledge_config.data_source.info_list.website_info_list  # type: ignore
                 if website_info:
                     count = len(website_info.urls)
             batch_upload_limit = int(dify_config.BATCH_UPLOAD_LIMIT)
@@ -1239,7 +1239,7 @@ class DocumentService:
         dataset = Dataset(
             tenant_id=tenant_id,
             name="",
-            data_source_type=knowledge_config.data_source.info_list.data_source_type,
+            data_source_type=knowledge_config.data_source.info_list.data_source_type,  # type: ignore
             indexing_technique=knowledge_config.indexing_technique,
             created_by=account.id,
             embedding_model=knowledge_config.embedding_model,
@@ -1611,8 +1611,11 @@ class SegmentService:
                     segment.answer = args.answer
                     segment.word_count += len(args.answer) if args.answer else 0
                 word_count_change = segment.word_count - word_count_change
+                keyword_changed = False
                 if args.keywords:
-                    segment.keywords = args.keywords
+                    if Counter(segment.keywords) != Counter(args.keywords):
+                        segment.keywords = args.keywords
+                        keyword_changed = True
                 segment.enabled = True
                 segment.disabled_at = None
                 segment.disabled_by = None
@@ -1623,13 +1626,6 @@ class SegmentService:
                     document.word_count = max(0, document.word_count + word_count_change)
                     db.session.add(document)
                 # update segment index task
-                if args.enabled:
-                    VectorService.create_segments_vector(
-                        [args.keywords] if args.keywords else None,
-                        [segment],
-                        dataset,
-                        document.doc_form,
-                    )
                 if document.doc_form == IndexType.PARENT_CHILD_INDEX and args.regenerate_child_chunks:
                     # regenerate child chunks
                     # get embedding model instance
@@ -1662,6 +1658,14 @@ class SegmentService:
                     VectorService.generate_child_chunks(
                         segment, document, dataset, embedding_model_instance, processing_rule, True
                     )
+                elif document.doc_form in (IndexType.PARAGRAPH_INDEX, IndexType.QA_INDEX):
+                    if args.enabled or keyword_changed:
+                        VectorService.create_segments_vector(
+                            [args.keywords] if args.keywords else None,
+                            [segment],
+                            dataset,
+                            document.doc_form,
+                        )
             else:
                 segment_hash = helper.generate_text_hash(content)
                 tokens = 0

api/services/entities/knowledge_entities/knowledge_entities.py

@@ -97,7 +97,7 @@ class KnowledgeConfig(BaseModel):
     original_document_id: Optional[str] = None
     duplicate: bool = True
     indexing_technique: Literal["high_quality", "economy"]
-    data_source: DataSource
+    data_source: Optional[DataSource] = None
     process_rule: Optional[ProcessRule] = None
     retrieval_model: Optional[RetrievalModel] = None
     doc_form: str = "text_model"

api/services/external_knowledge_service.py

@@ -155,7 +155,7 @@ class ExternalDatasetService:
             if custom_parameters:
                 for parameter in custom_parameters:
                     if parameter.get("required", False) and not process_parameter.get(parameter.get("name")):
-                        raise ValueError(f'{parameter.get("name")} is required')
+                        raise ValueError(f"{parameter.get('name')} is required")
 
     @staticmethod
     def process_external_api(

api/tasks/batch_create_segment_to_index_task.py

@@ -5,7 +5,8 @@ import uuid
 
 import click
 from celery import shared_task  # type: ignore
-from sqlalchemy import func
+from sqlalchemy import func, select
+from sqlalchemy.orm import Session
 
 from core.model_manager import ModelManager
 from core.model_runtime.entities.model_entities import ModelType
@@ -18,7 +19,12 @@ from services.vector_service import VectorService
 
 @shared_task(queue="dataset")
 def batch_create_segment_to_index_task(
-    job_id: str, content: list, dataset_id: str, document_id: str, tenant_id: str, user_id: str
+    job_id: str,
+    content: list,
+    dataset_id: str,
+    document_id: str,
+    tenant_id: str,
+    user_id: str,
 ):
     """
     Async batch create segment to index
@@ -37,71 +43,80 @@ def batch_create_segment_to_index_task(
     indexing_cache_key = "segment_batch_import_{}".format(job_id)
 
     try:
-        dataset = db.session.query(Dataset).filter(Dataset.id == dataset_id).first()
-        if not dataset:
-            raise ValueError("Dataset not exist.")
+        with Session(db.engine) as session:
+            dataset = session.get(Dataset, dataset_id)
+            if not dataset:
+                raise ValueError("Dataset not exist.")
 
-        dataset_document = db.session.query(Document).filter(Document.id == document_id).first()
-        if not dataset_document:
-            raise ValueError("Document not exist.")
+            dataset_document = session.get(Document, document_id)
+            if not dataset_document:
+                raise ValueError("Document not exist.")
 
-        if not dataset_document.enabled or dataset_document.archived or dataset_document.indexing_status != "completed":
-            raise ValueError("Document is not available.")
-        document_segments = []
-        embedding_model = None
-        if dataset.indexing_technique == "high_quality":
-            model_manager = ModelManager()
-            embedding_model = model_manager.get_model_instance(
-                tenant_id=dataset.tenant_id,
-                provider=dataset.embedding_model_provider,
-                model_type=ModelType.TEXT_EMBEDDING,
-                model=dataset.embedding_model,
+            if (
+                not dataset_document.enabled
+                or dataset_document.archived
+                or dataset_document.indexing_status != "completed"
+            ):
+                raise ValueError("Document is not available.")
+            document_segments = []
+            embedding_model = None
+            if dataset.indexing_technique == "high_quality":
+                model_manager = ModelManager()
+                embedding_model = model_manager.get_model_instance(
+                    tenant_id=dataset.tenant_id,
+                    provider=dataset.embedding_model_provider,
+                    model_type=ModelType.TEXT_EMBEDDING,
+                    model=dataset.embedding_model,
+                )
+            word_count_change = 0
+            segments_to_insert: list[str] = []
+            max_position_stmt = select(func.max(DocumentSegment.position)).where(
+                DocumentSegment.document_id == dataset_document.id
             )
-        word_count_change = 0
-        segments_to_insert: list[str] = []  # Explicitly type hint the list as List[str]
-        for segment in content:
-            content_str = segment["content"]
-            doc_id = str(uuid.uuid4())
-            segment_hash = helper.generate_text_hash(content_str)
-            # calc embedding use tokens
-            tokens = embedding_model.get_text_embedding_num_tokens(texts=[content_str]) if embedding_model else 0
-            max_position = (
-                db.session.query(func.max(DocumentSegment.position))
-                .filter(DocumentSegment.document_id == dataset_document.id)
-                .scalar()
-            )
-            segment_document = DocumentSegment(
-                tenant_id=tenant_id,
-                dataset_id=dataset_id,
-                document_id=document_id,
-                index_node_id=doc_id,
-                index_node_hash=segment_hash,
-                position=max_position + 1 if max_position else 1,
-                content=content_str,
-                word_count=len(content_str),
-                tokens=tokens,
-                created_by=user_id,
-                indexing_at=datetime.datetime.now(datetime.UTC).replace(tzinfo=None),
-                status="completed",
-                completed_at=datetime.datetime.now(datetime.UTC).replace(tzinfo=None),
-            )
-            if dataset_document.doc_form == "qa_model":
-                segment_document.answer = segment["answer"]
-                segment_document.word_count += len(segment["answer"])
-            word_count_change += segment_document.word_count
-            db.session.add(segment_document)
-            document_segments.append(segment_document)
-            segments_to_insert.append(str(segment))  # Cast to string if needed
-        # update document word count
-        dataset_document.word_count += word_count_change
-        db.session.add(dataset_document)
-        # add index to db
-        VectorService.create_segments_vector(None, document_segments, dataset, dataset_document.doc_form)
-        db.session.commit()
+            max_position = session.scalar(max_position_stmt) or 1
+            for segment in content:
+                content_str = segment["content"]
+                doc_id = str(uuid.uuid4())
+                segment_hash = helper.generate_text_hash(content_str)
+                # calc embedding use tokens
+                tokens = embedding_model.get_text_embedding_num_tokens(texts=[content_str]) if embedding_model else 0
+                segment_document = DocumentSegment(
+                    tenant_id=tenant_id,
+                    dataset_id=dataset_id,
+                    document_id=document_id,
+                    index_node_id=doc_id,
+                    index_node_hash=segment_hash,
+                    position=max_position,
+                    content=content_str,
+                    word_count=len(content_str),
+                    tokens=tokens,
+                    created_by=user_id,
+                    indexing_at=datetime.datetime.now(datetime.UTC).replace(tzinfo=None),
+                    status="completed",
+                    completed_at=datetime.datetime.now(datetime.UTC).replace(tzinfo=None),
+                )
+                max_position += 1
+                if dataset_document.doc_form == "qa_model":
+                    segment_document.answer = segment["answer"]
+                    segment_document.word_count += len(segment["answer"])
+                word_count_change += segment_document.word_count
+                session.add(segment_document)
+                document_segments.append(segment_document)
+                segments_to_insert.append(str(segment))  # Cast to string if needed
+            # update document word count
+            dataset_document.word_count += word_count_change
+            session.add(dataset_document)
+            # add index to db
+            VectorService.create_segments_vector(None, document_segments, dataset, dataset_document.doc_form)
+            session.commit()
+
         redis_client.setex(indexing_cache_key, 600, "completed")
         end_at = time.perf_counter()
         logging.info(
-            click.style("Segment batch created job: {} latency: {}".format(job_id, end_at - start_at), fg="green")
+            click.style(
+                "Segment batch created job: {} latency: {}".format(job_id, end_at - start_at),
+                fg="green",
+            )
         )
     except Exception as e:
         logging.exception("Segments batch created index failed")

api/tests/artifact_tests/dependencies/test_dependencies_sorted.py

@@ -44,6 +44,6 @@ def test_duplicated_dependency_crossing_groups() -> None:
         dependency_names = list(dependencies.keys())
         all_dependency_names.extend(dependency_names)
     expected_all_dependency_names = set(all_dependency_names)
-    assert sorted(expected_all_dependency_names) == sorted(
-        all_dependency_names
-    ), "Duplicated dependencies crossing groups are found"
+    assert sorted(expected_all_dependency_names) == sorted(all_dependency_names), (
+        "Duplicated dependencies crossing groups are found"
+    )

api/tests/integration_tests/vdb/opensearch/test_opensearch.py

@@ -89,9 +89,9 @@ class TestOpenSearchVector:
         print("Actual document ID:", hits_by_vector[0].metadata["document_id"] if hits_by_vector else "No hits")
 
         assert len(hits_by_vector) > 0, f"Expected at least one hit, got {len(hits_by_vector)}"
-        assert (
-            hits_by_vector[0].metadata["document_id"] == self.example_doc_id
-        ), f"Expected document ID {self.example_doc_id}, got {hits_by_vector[0].metadata['document_id']}"
+        assert hits_by_vector[0].metadata["document_id"] == self.example_doc_id, (
+            f"Expected document ID {self.example_doc_id}, got {hits_by_vector[0].metadata['document_id']}"
+        )
 
     def test_get_ids_by_metadata_field(self):
         mock_response = {"hits": {"total": {"value": 1}, "hits": [{"_id": "mock_id"}]}}

api/tests/unit_tests/core/workflow/nodes/llm/test_node.py

@@ -438,9 +438,9 @@ def test_fetch_prompt_messages__basic(faker, llm_node, model_config):
 
         # Verify the result
         assert len(prompt_messages) == len(scenario.expected_messages), f"Scenario failed: {scenario.description}"
-        assert (
-            prompt_messages == scenario.expected_messages
-        ), f"Message content mismatch in scenario: {scenario.description}"
+        assert prompt_messages == scenario.expected_messages, (
+            f"Message content mismatch in scenario: {scenario.description}"
+        )
 
 
 def test_handle_list_messages_basic(llm_node):

api/tests/unit_tests/services/workflow/test_workflow_converter.py

@@ -401,8 +401,7 @@ def test__convert_to_llm_node_for_workflow_advanced_completion_model(default_var
     prompt_template = PromptTemplateEntity(
         prompt_type=PromptTemplateEntity.PromptType.ADVANCED,
         advanced_completion_prompt_template=AdvancedCompletionPromptTemplateEntity(
-            prompt="You are a helpful assistant named {{name}}.\n\nContext:\n{{#context#}}\n\n"
-            "Human: hi\nAssistant: ",
+            prompt="You are a helpful assistant named {{name}}.\n\nContext:\n{{#context#}}\n\nHuman: hi\nAssistant: ",
             role_prefix=AdvancedCompletionPromptTemplateEntity.RolePrefixEntity(user="Human", assistant="Assistant"),
         ),
     )

dev/reformat

@@ -9,10 +9,10 @@ if ! command -v ruff &> /dev/null || ! command -v dotenv-linter &> /dev/null; th
 fi
 
 # run ruff linter
-poetry run -C api ruff check --fix ./api
+poetry run -C api ruff check --fix ./
 
 # run ruff formatter
-poetry run -C api ruff format ./api
+poetry run -C api ruff format ./
 
 # run dotenv-linter linter
-poetry run -C api dotenv-linter ./api/.env.example ./web/.env.example
+poetry run -P api dotenv-linter ./api/.env.example ./web/.env.example

dev/sync-poetry

@@ -12,7 +12,7 @@ if [ $? -ne 0 ]; then
     # update poetry.lock
     # refreshing lockfile only without updating locked versions
     echo "poetry.lock is outdated, refreshing without updating locked versions ..."
-    poetry lock -C api --no-update
+    poetry lock -C api
 else
   echo "poetry.lock is ready."
 fi

docker-legacy/docker-compose.yaml

@@ -2,7 +2,7 @@ version: '3'
 services:
   # API service
   api:
-    image: langgenius/dify-api:0.15.1
+    image: langgenius/dify-api:0.15.2
     restart: always
     environment:
       # Startup mode, 'api' starts the API server.
@@ -227,7 +227,7 @@ services:
   # worker service
   # The Celery worker for processing the queue.
   worker:
-    image: langgenius/dify-api:0.15.1
+    image: langgenius/dify-api:0.15.2
     restart: always
     environment:
       CONSOLE_WEB_URL: ''
@@ -397,7 +397,7 @@ services:
 
   # Frontend web application.
   web:
-    image: langgenius/dify-web:0.15.1
+    image: langgenius/dify-web:0.15.2
     restart: always
     environment:
       # The base URL of console application api server, refers to the Console base URL of WEB service if console domain is

docker/docker-compose-template.yaml

@@ -2,7 +2,7 @@ x-shared-env: &shared-api-worker-env
 services:
   # API service
   api:
-    image: langgenius/dify-api:0.15.1
+    image: langgenius/dify-api:0.15.2
     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:0.15.1
+    image: langgenius/dify-api:0.15.2
     restart: always
     environment:
       # Use the shared environment variables.
@@ -47,7 +47,7 @@ services:
 
   # Frontend web application.
   web:
-    image: langgenius/dify-web:0.15.1
+    image: langgenius/dify-web:0.15.2
     restart: always
     environment:
       CONSOLE_API_URL: ${CONSOLE_API_URL:-}
@@ -57,6 +57,7 @@ services:
       TEXT_GENERATION_TIMEOUT_MS: ${TEXT_GENERATION_TIMEOUT_MS:-60000}
       CSP_WHITELIST: ${CSP_WHITELIST:-}
       TOP_K_MAX_VALUE: ${TOP_K_MAX_VALUE:-}
+      INDEXING_MAX_SEGMENTATION_TOKENS_LENGTH: ${INDEXING_MAX_SEGMENTATION_TOKENS_LENGTH:-}
 
   # The postgres database.
   db:

docker/docker-compose.yaml

@@ -393,7 +393,7 @@ x-shared-env: &shared-api-worker-env
 services:
   # API service
   api:
-    image: langgenius/dify-api:0.15.1
+    image: langgenius/dify-api:0.15.2
     restart: always
     environment:
       # Use the shared environment variables.
@@ -416,7 +416,7 @@ services:
   # worker service
   # The Celery worker for processing the queue.
   worker:
-    image: langgenius/dify-api:0.15.1
+    image: langgenius/dify-api:0.15.2
     restart: always
     environment:
       # Use the shared environment variables.
@@ -438,7 +438,7 @@ services:
 
   # Frontend web application.
   web:
-    image: langgenius/dify-web:0.15.1
+    image: langgenius/dify-web:0.15.2
     restart: always
     environment:
       CONSOLE_API_URL: ${CONSOLE_API_URL:-}
@@ -448,6 +448,7 @@ services:
       TEXT_GENERATION_TIMEOUT_MS: ${TEXT_GENERATION_TIMEOUT_MS:-60000}
       CSP_WHITELIST: ${CSP_WHITELIST:-}
       TOP_K_MAX_VALUE: ${TOP_K_MAX_VALUE:-}
+      INDEXING_MAX_SEGMENTATION_TOKENS_LENGTH: ${INDEXING_MAX_SEGMENTATION_TOKENS_LENGTH:-}
 
   # The postgres database.
   db:

web/.env.example

@@ -28,3 +28,6 @@ NEXT_PUBLIC_CSP_WHITELIST=
 
 # The maximum number of top-k value for RAG.
 NEXT_PUBLIC_TOP_K_MAX_VALUE=10
+
+# The maximum number of tokens for segmentation
+NEXT_PUBLIC_INDEXING_MAX_SEGMENTATION_TOKENS_LENGTH=4000

web/app/(shareLayout)/layout.tsx

@@ -1,7 +1,6 @@
 import React from 'react'
 import type { FC } from 'react'
 import type { Metadata } from 'next'
-import GA, { GaType } from '@/app/components/base/ga'
 
 export const metadata: Metadata = {
   icons: 'data:,', // prevent browser from using default favicon
@@ -12,7 +11,6 @@ const Layout: FC<{
 }> = ({ children }) => {
   return (
     <div className="min-w-[300px] h-full pb-[env(safe-area-inset-bottom)]">
-      <GA gaType={GaType.webapp} />
       {children}
     </div>
   )

web/app/account/account-page/AvatarWithEdit.tsx

@@ -0,0 +1,122 @@
+'use client'
+
+import type { Area } from 'react-easy-crop'
+import React, { useCallback, useState } from 'react'
+import { useTranslation } from 'react-i18next'
+import { useContext } from 'use-context-selector'
+import { RiPencilLine } from '@remixicon/react'
+import { updateUserProfile } from '@/service/common'
+import { ToastContext } from '@/app/components/base/toast'
+import ImageInput, { type OnImageInput } from '@/app/components/base/app-icon-picker/ImageInput'
+import Modal from '@/app/components/base/modal'
+import Divider from '@/app/components/base/divider'
+import Button from '@/app/components/base/button'
+import Avatar, { type AvatarProps } from '@/app/components/base/avatar'
+import { useLocalFileUploader } from '@/app/components/base/image-uploader/hooks'
+import type { ImageFile } from '@/types/app'
+import getCroppedImg from '@/app/components/base/app-icon-picker/utils'
+import { DISABLE_UPLOAD_IMAGE_AS_ICON } from '@/config'
+
+type InputImageInfo = { file: File } | { tempUrl: string; croppedAreaPixels: Area; fileName: string }
+type AvatarWithEditProps = AvatarProps & { onSave?: () => void }
+
+const AvatarWithEdit = ({ onSave, ...props }: AvatarWithEditProps) => {
+  const { t } = useTranslation()
+  const { notify } = useContext(ToastContext)
+
+  const [inputImageInfo, setInputImageInfo] = useState<InputImageInfo>()
+  const [isShowAvatarPicker, setIsShowAvatarPicker] = useState(false)
+  const [uploading, setUploading] = useState(false)
+
+  const handleImageInput: OnImageInput = useCallback(async (isCropped: boolean, fileOrTempUrl: string | File, croppedAreaPixels?: Area, fileName?: string) => {
+    setInputImageInfo(
+      isCropped
+        ? { tempUrl: fileOrTempUrl as string, croppedAreaPixels: croppedAreaPixels!, fileName: fileName! }
+        : { file: fileOrTempUrl as File },
+    )
+  }, [setInputImageInfo])
+
+  const handleSaveAvatar = useCallback(async (uploadedFileId: string) => {
+    try {
+      await updateUserProfile({ url: 'account/avatar', body: { avatar: uploadedFileId } })
+      notify({ type: 'success', message: t('common.actionMsg.modifiedSuccessfully') })
+      setIsShowAvatarPicker(false)
+      onSave?.()
+    }
+    catch (e) {
+      notify({ type: 'error', message: (e as Error).message })
+    }
+  }, [notify, onSave, t])
+
+  const { handleLocalFileUpload } = useLocalFileUploader({
+    limit: 3,
+    disabled: false,
+    onUpload: (imageFile: ImageFile) => {
+      if (imageFile.progress === 100) {
+        setUploading(false)
+        setInputImageInfo(undefined)
+        handleSaveAvatar(imageFile.fileId)
+      }
+
+      // Error
+      if (imageFile.progress === -1)
+        setUploading(false)
+    },
+  })
+
+  const handleSelect = useCallback(async () => {
+    if (!inputImageInfo)
+      return
+    setUploading(true)
+    if ('file' in inputImageInfo) {
+      handleLocalFileUpload(inputImageInfo.file)
+      return
+    }
+    const blob = await getCroppedImg(inputImageInfo.tempUrl, inputImageInfo.croppedAreaPixels, inputImageInfo.fileName)
+    const file = new File([blob], inputImageInfo.fileName, { type: blob.type })
+    handleLocalFileUpload(file)
+  }, [handleLocalFileUpload, inputImageInfo])
+
+  if (DISABLE_UPLOAD_IMAGE_AS_ICON)
+    return <Avatar {...props} />
+
+  return (
+    <>
+      <div>
+        <div className="relative group">
+          <Avatar {...props} />
+          <div
+            onClick={() => { setIsShowAvatarPicker(true) }}
+            className="absolute inset-0 bg-black bg-opacity-50 rounded-full opacity-0 group-hover:opacity-100 transition-opacity cursor-pointer flex items-center justify-center"
+          >
+            <span className="text-white text-xs">
+              <RiPencilLine />
+            </span>
+          </div>
+        </div>
+      </div>
+
+      <Modal
+        closable
+        className="!w-[362px] !p-0"
+        isShow={isShowAvatarPicker}
+        onClose={() => setIsShowAvatarPicker(false)}
+      >
+        <ImageInput onImageInput={handleImageInput} cropShape='round' />
+        <Divider className='m-0' />
+
+        <div className='w-full flex items-center justify-center p-3 gap-2'>
+          <Button className='w-full' onClick={() => setIsShowAvatarPicker(false)}>
+            {t('app.iconPicker.cancel')}
+          </Button>
+
+          <Button variant="primary" className='w-full' disabled={uploading || !inputImageInfo} loading={uploading} onClick={handleSelect}>
+            {t('app.iconPicker.ok')}
+          </Button>
+        </div>
+      </Modal>
+    </>
+  )
+}
+
+export default AvatarWithEdit

web/app/account/account-page/index.tsx

@@ -5,6 +5,7 @@ import { useTranslation } from 'react-i18next'
 import { useContext } from 'use-context-selector'
 import DeleteAccount from '../delete-account'
 import s from './index.module.css'
+import AvatarWithEdit from './AvatarWithEdit'
 import Collapse from '@/app/components/header/account-setting/collapse'
 import type { IItem } from '@/app/components/header/account-setting/collapse'
 import Modal from '@/app/components/base/modal'
@@ -13,7 +14,6 @@ import { updateUserProfile } from '@/service/common'
 import { useAppContext } from '@/context/app-context'
 import { ToastContext } from '@/app/components/base/toast'
 import AppIcon from '@/app/components/base/app-icon'
-import Avatar from '@/app/components/base/avatar'
 import { IS_CE_EDITION } from '@/config'
 import Input from '@/app/components/base/input'
 
@@ -133,7 +133,7 @@ export default function AccountPage() {
         <h4 className='title-2xl-semi-bold text-text-primary'>{t('common.account.myAccount')}</h4>
       </div>
       <div className='mb-8 p-6 rounded-xl flex items-center bg-gradient-to-r from-background-gradient-bg-fill-chat-bg-2 to-background-gradient-bg-fill-chat-bg-1'>
-        <Avatar name={userProfile.name} size={64} />
+        <AvatarWithEdit avatar={userProfile.avatar_url} name={userProfile.name} onSave={ mutateUserProfile } size={64} />
         <div className='ml-4'>
           <p className='system-xl-semibold text-text-primary'>{userProfile.name}</p>
           <p className='system-xs-regular text-text-tertiary'>{userProfile.email}</p>

web/app/account/avatar.tsx

@@ -45,7 +45,7 @@ export default function AppSelector() {
                     ${open && 'bg-components-panel-bg-blur'}
                   `}
               >
-                <Avatar name={userProfile.name} size={32} />
+                <Avatar avatar={userProfile.avatar_url} name={userProfile.name} size={32} />
               </Menu.Button>
             </div>
             <Transition
@@ -71,7 +71,7 @@ export default function AppSelector() {
                         <div className='system-md-medium text-text-primary break-all'>{userProfile.name}</div>
                         <div className='system-xs-regular text-text-tertiary break-all'>{userProfile.email}</div>
                       </div>
-                      <Avatar name={userProfile.name} size={32} />
+                      <Avatar avatar={userProfile.avatar_url} name={userProfile.name} size={32} />
                     </div>
                   </div>
                 </Menu.Item>

web/app/components/app/configuration/debug/debug-with-multiple-model/chat-item.tsx

@@ -149,7 +149,7 @@ const ChatItem: FC<ChatItemProps> = ({
       suggestedQuestions={suggestedQuestions}
       onSend={doSend}
       showPromptLog
-      questionIcon={<Avatar name={userProfile.name} size={40} />}
+      questionIcon={<Avatar avatar={userProfile.avatar_url} name={userProfile.name} size={40} />}
       allToolIcons={allToolIcons}
       hideLogModal
       noSpacing

web/app/components/app/configuration/debug/debug-with-single-model/index.tsx

@@ -175,7 +175,7 @@ const DebugWithSingleModel = forwardRef<DebugWithSingleModelRefType, DebugWithSi
       onRegenerate={doRegenerate}
       onStopResponding={handleStop}
       showPromptLog
-      questionIcon={<Avatar name={userProfile.name} size={40} />}
+      questionIcon={<Avatar avatar={userProfile.avatar_url} name={userProfile.name} size={40} />}
       allToolIcons={allToolIcons}
       onAnnotationEdited={handleAnnotationEdited}
       onAnnotationAdded={handleAnnotationAdded}

web/app/components/app/log/var-panel.tsx

@@ -39,7 +39,7 @@ const VarPanel: FC<Props> = ({
         }
       </div>
       {!isCollapse && (
-        <div className='p-3 flex flex-col gap-2'>
+        <div className='p-3 flex flex-col gap-2 max-h-[500px] overflow-y-auto'>
           {varList.map(({ label, value }, index) => (
             <div key={index} className='flex py-2 system-xs-medium'>
               <div className='shrink-0 w-[128px] flex text-text-accent'>