langchain-dev-utils 1.1.12__tar.gz → 1.1.14__tar.gz

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: langchain-dev-utils
-Version: 1.1.12
+Version: 1.1.14
 Summary: A practical utility library for LangChain and LangGraph development
 Project-URL: Source Code, https://github.com/TBice123123/langchain-dev-utils
 Project-URL: repository, https://github.com/TBice123123/langchain-dev-utils
@@ -61,7 +61,7 @@ Mainly consists of the following two functions:
 
 - `provider_name`: Model provider name, used as an identifier for subsequent model loading
 - `chat_model`: Chat model, can be a ChatModel or a string (currently supports "openai-compatible")
-- `base_url`: API address of the model provider (optional, valid when `chat_model` is a string and is "openai-compatible")
+- `base_url`: The API address of the model provider (optional, valid for both types of `chat_model`, but mainly used when `chat_model` is a string and is "openai-compatible")
 - `provider_config`: Relevant configuration for the model provider (optional, valid when `chat_model` is a string and is "openai-compatible"), can configure some provider-related parameters, such as whether to support structured output in json_mode, list of supported tool_choices, etc.
 
 `load_chat_model` parameter description:
@@ -101,7 +101,7 @@ Mainly consists of the following two functions:
 
 - `provider_name`: Embedding model provider name, used as an identifier for subsequent model loading
 - `embeddings_model`: Embedding model, can be Embeddings or a string (currently supports "openai-compatible")
-- `base_url`: API address of the model provider (optional, valid when `embeddings_model` is a string and is "openai-compatible")
+- `base_url`: The API address of the Embedding model provider (optional, valid for both types of `embeddings_model`, but mainly used when `embeddings_model` is a string and is "openai-compatible")
 
 `load_embeddings` parameter description:
 

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/README.md

@@ -45,7 +45,7 @@ Mainly consists of the following two functions:
 
 - `provider_name`: Model provider name, used as an identifier for subsequent model loading
 - `chat_model`: Chat model, can be a ChatModel or a string (currently supports "openai-compatible")
-- `base_url`: API address of the model provider (optional, valid when `chat_model` is a string and is "openai-compatible")
+- `base_url`: The API address of the model provider (optional, valid for both types of `chat_model`, but mainly used when `chat_model` is a string and is "openai-compatible")
 - `provider_config`: Relevant configuration for the model provider (optional, valid when `chat_model` is a string and is "openai-compatible"), can configure some provider-related parameters, such as whether to support structured output in json_mode, list of supported tool_choices, etc.
 
 `load_chat_model` parameter description:
@@ -85,7 +85,7 @@ Mainly consists of the following two functions:
 
 - `provider_name`: Embedding model provider name, used as an identifier for subsequent model loading
 - `embeddings_model`: Embedding model, can be Embeddings or a string (currently supports "openai-compatible")
-- `base_url`: API address of the model provider (optional, valid when `embeddings_model` is a string and is "openai-compatible")
+- `base_url`: The API address of the Embedding model provider (optional, valid for both types of `embeddings_model`, but mainly used when `embeddings_model` is a string and is "openai-compatible")
 
 `load_embeddings` parameter description:
 

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/README_cn.md

@@ -45,7 +45,7 @@ pip install -U langchain-dev-utils[standard]
 
 - `provider_name`：模型提供商名称，作为后续模型加载的标识
 - `chat_model`：对话模型，可以是 ChatModel 或字符串（目前支持 "openai-compatible"）
-- `base_url`：模型提供商的 API 地址（可选，当 `chat_model` 为字符串且是"openai-compatible"时有效）
+- `base_url`：模型提供商的 API 地址（可选，对于`chat_model`的两种类型情况都有效，但是主要用于`chat_model`为字符串且是"openai-compatible"的情况）
 - `provider_config`：模型提供商的相关配置（可选，当 `chat_model` 为字符串且是 "openai-compatible" 时有效），可以配置一些提供商的相关参数，例如是否支持 json_mode 的结构化输出方式、支持的 tool_choice 列表等
 
 `load_chat_model` 参数说明：
@@ -85,7 +85,7 @@ print(model.invoke("Hello"))
 
 - `provider_name`：嵌入模型提供商名称，作为后续模型加载的标识
 - `embeddings_model`：嵌入模型，可以是 Embeddings 或字符串（目前支持 "openai-compatible"）
-- `base_url`：模型提供商的 API 地址（可选，当 `embeddings_model` 为字符串且是"openai-compatible"时有效）
+- `base_url`：嵌入模型提供商的 API 地址（可选，对于`embeddings_model`的两种类型情况都有效，但是主要用于`embeddings_model`为字符串且是"openai-compatible"的情况）
 
 `load_embeddings` 参数说明：
 

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "langchain-dev-utils"
-version = "1.1.12"
+version = "1.1.14"
 description = "A practical utility library for LangChain and LangGraph development"
 readme = "README.md"
 authors = [{ name = "tiebingice", email = "tiebingice123@outlook.com" }]
@@ -28,8 +28,10 @@ python_files = ["test_*.py"]
 python_functions = ["test_*"]
 
 [dependency-groups]
+dev = ["ruff>=0.14.5"]
 tests = [
     "python-dotenv>=1.1.1",
     "langchain-tests>=1.0.0",
     "langchain-deepseek>=1.0.0",
+    "langchain-qwq>=0.3.0",
 ]

langchain_dev_utils-1.1.14/src/langchain_dev_utils/__init__.py

@@ -0,0 +1 @@
+__version__ = "1.1.14"

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/src/langchain_dev_utils/agents/factory.py

@@ -1,9 +1,14 @@
 from typing import Any, Sequence
-from typing import Callable, cast
+from typing import Callable
 
-from langchain.agents import AgentState, create_agent as _create_agent
-from langchain.agents.factory import ResponseT
-from langchain.agents.middleware import AgentMiddleware
+from langchain.agents import create_agent as _create_agent
+from langchain.agents.middleware.types import (
+    AgentMiddleware,
+    AgentState,
+    ResponseT,
+    _InputAgentState,
+    _OutputAgentState,
+)
 from langchain.agents.structured_output import ResponseFormat
 from langchain_core.tools import BaseTool
 from langgraph.cache.base import BaseCache
@@ -31,7 +36,9 @@ def create_agent(  # noqa: PLR0915
     debug: bool = False,
     name: str | None = None,
     cache: BaseCache | None = None,
-) -> CompiledStateGraph:
+) -> CompiledStateGraph[
+    AgentState[ResponseT], ContextT, _InputAgentState, _OutputAgentState[ResponseT]
+]:
     """
     Create a prebuilt agent with string-based model specification.
 
@@ -80,22 +87,19 @@ def create_agent(  # noqa: PLR0915
         ... })
         >>> response
     """
-    return cast(
-        CompiledStateGraph,
-        _create_agent(
-            model=load_chat_model(model),
-            tools=tools,
-            system_prompt=system_prompt,
-            middleware=middleware,
-            response_format=response_format,
-            state_schema=state_schema,
-            context_schema=context_schema,
-            checkpointer=checkpointer,
-            store=store,
-            interrupt_before=interrupt_before,
-            interrupt_after=interrupt_after,
-            debug=debug,
-            name=name,
-            cache=cache,
-        ),
+    return _create_agent(
+        model=load_chat_model(model),
+        tools=tools,
+        system_prompt=system_prompt,
+        middleware=middleware,
+        response_format=response_format,
+        state_schema=state_schema,
+        context_schema=context_schema,
+        checkpointer=checkpointer,
+        store=store,
+        interrupt_before=interrupt_before,
+        interrupt_after=interrupt_after,
+        debug=debug,
+        name=name,
+        cache=cache,
     )

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/src/langchain_dev_utils/agents/middleware/plan.py

@@ -244,9 +244,10 @@ def create_read_plan_tool(
 
 _PLAN_SYSTEM_PROMPT_NOT_READ_PLAN = """You can manage task plans using two simple tools:
 
-1. **write_plan**
+## write_plan
 - Use it to break complex tasks (3+ steps) into a clear, actionable list. Only include next steps to execute — the first becomes `"in_progress"`, the rest `"pending"`. Don’t use it for simple tasks (<3 steps).
-2. **finish_sub_plan**
+
+## finish_sub_plan
 - Call it **only when the current task is 100% done**. It automatically marks it `"done"` and promotes the next `"pending"` task to `"in_progress"`. No parameters needed. Never use it mid-task or if anything’s incomplete.
 Keep plans lean, update immediately, and never batch completions.
 """
@@ -277,9 +278,16 @@ class PlanMiddleware(AgentMiddleware):
 
     Args:
         system_prompt: Custom system prompt to guide the agent on using the plan tool.
-            If not provided, uses the default `_PLAN_MIDDLEWARE_SYSTEM_PROMPT`.
-        tools: List of tools to be added to the agent. The tools must be created by `create_write_plan_tool`, `create_finish_sub_plan_tool`, and `create_read_plan_tool`(optional).
-
+            If not provided, uses the default `_PLAN_SYSTEM_PROMPT` or `_PLAN_SYSTEM_PROMPT_NOT_READ_PLAN` based on the `use_read_plan_tool` parameter.
+        write_plan_tool_description: Description of the `write_plan` tool.
+            If not provided, uses the default `_DEFAULT_WRITE_PLAN_TOOL_DESCRIPTION`.
+        finish_sub_plan_tool_description: Description of the `finish_sub_plan` tool.
+            If not provided, uses the default `_DEFAULT_FINISH_SUB_PLAN_TOOL_DESCRIPTION`.
+        read_plan_tool_description: Description of the `read_plan` tool.
+            If not provided, uses the default `_DEFAULT_READ_PLAN_TOOL_DESCRIPTION`.
+        use_read_plan_tool: Whether to use the `read_plan` tool.
+            If not provided, uses the default `True`.
+        message_key: The key of the message to be updated. Defaults to "messages".
     Example:
         ```python
         from langchain_dev_utils.agents.middleware.plan import PlanMiddleware
@@ -304,6 +312,7 @@ class PlanMiddleware(AgentMiddleware):
         finish_sub_plan_tool_description: Optional[str] = None,
         read_plan_tool_description: Optional[str] = None,
         use_read_plan_tool: bool = True,
+        message_key: Optional[str] = None,
     ) -> None:
         super().__init__()
 
@@ -319,8 +328,12 @@ class PlanMiddleware(AgentMiddleware):
         )
 
         tools = [
-            create_write_plan_tool(description=write_plan_tool_description),
-            create_finish_sub_plan_tool(description=finish_sub_plan_tool_description),
+            create_write_plan_tool(
+                description=write_plan_tool_description, message_key=message_key
+            ),
+            create_finish_sub_plan_tool(
+                description=finish_sub_plan_tool_description, message_key=message_key
+            ),
         ]
 
         if use_read_plan_tool:

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/src/langchain_dev_utils/chat_models/adapters/openai_compatible.py

@@ -46,7 +46,7 @@ _DictOrPydantic = Union[dict, _BM]
 
 
 class _ModelProviderConfigType(BaseModel):
-    supported_tool_choice: ToolChoiceType = Field(default=[])
+    supported_tool_choice: ToolChoiceType = Field(default_factory=list)
     keep_reasoning_content: bool = Field(default=False)
     support_json_mode: bool = Field(default=False)
 

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/src/langchain_dev_utils/chat_models/base.py

@@ -1,8 +1,9 @@
-import os
 from typing import Any, NotRequired, Optional, TypedDict, cast
 
 from langchain.chat_models.base import _SUPPORTED_PROVIDERS, _init_chat_model_helper
 from langchain_core.language_models.chat_models import BaseChatModel
+from langchain_core.utils import from_env
+from pydantic import BaseModel
 
 from .types import ChatModelType, ToolChoiceType
 
@@ -22,6 +23,34 @@ class ChatModelProvider(TypedDict):
     provider_config: NotRequired[ProviderConfig]
 
 
+def _get_base_url_field_name(model_cls: type[BaseModel]) -> str | None:
+    """
+    Return 'base_url' if the model has a field named or aliased as 'base_url',
+    else return 'api_base' if it has a field named or aliased as 'api_base',
+    else return None.
+    The return value is always either 'base_url', 'api_base', or None.
+    """
+    model_fields = model_cls.model_fields
+
+    # try model_fields first
+    if "base_url" in model_fields:
+        return "base_url"
+
+    if "api_base" in model_fields:
+        return "api_base"
+
+    # then try aliases
+    for field_info in model_fields.values():
+        if field_info.alias == "base_url":
+            return "base_url"
+
+    for field_info in model_fields.values():
+        if field_info.alias == "api_base":
+            return "api_base"
+
+    return None
+
+
 def _parse_model(model: str, model_provider: Optional[str]) -> tuple[str, str]:
     """Parse model string and provider.
 
@@ -71,6 +100,11 @@ def _load_chat_model_helper(
             "provider_config"
         ):
             kwargs.update({"provider_config": provider_config})
+
+        if base_url := _MODEL_PROVIDERS_DICT[model_provider].get("base_url"):
+            url_key = _get_base_url_field_name(chat_model)
+            if url_key:
+                kwargs.update({url_key: base_url})
         return chat_model(model=model, **kwargs)
 
     return _init_chat_model_helper(model, model_provider=model_provider, **kwargs)
@@ -91,7 +125,7 @@ def register_model_provider(
     Args:
         provider_name: Name of the provider to register
         chat_model: Either a BaseChatModel class or a string identifier for a supported provider
-        base_url: Optional base URL for API endpoints (Optional parameter;effective only when `chat_model` is a string and is "openai-compatible".)
+        base_url: The API address of the model provider (optional, valid for both types of `chat_model`, but mainly used when `chat_model` is a string and is "openai-compatible")
         provider_config: The configuration of the model provider (Optional parameter;effective only when `chat_model` is a string and is "openai-compatible".)
            It can be configured to configure some related parameters of the provider, such as whether to support json_mode structured output mode, the list of supported tool_choice
     Raises:
@@ -113,6 +147,7 @@ def register_model_provider(
         >>> model = load_chat_model(model="vllm:qwen3-4b")
         >>> model.invoke("Hello")
     """
+    base_url = base_url or from_env(f"{provider_name.upper()}_API_BASE", default=None)()
     if isinstance(chat_model, str):
         try:
             from .adapters.openai_compatible import _create_openai_compatible_model
@@ -120,8 +155,6 @@ def register_model_provider(
             raise ImportError(
                 "Please install langchain_dev_utils[standard],when chat_model is a 'openai-compatible'"
             )
-
-        base_url = base_url or os.getenv(f"{provider_name.upper()}_API_BASE")
         if base_url is None:
             raise ValueError(
                 f"base_url must be provided or set {provider_name.upper()}_API_BASE environment variable when chat_model is a string"
@@ -140,11 +173,17 @@ def register_model_provider(
                 provider_name: {
                     "chat_model": chat_model,
                     "provider_config": provider_config,
+                    "base_url": base_url,
                 }
             }
         )
     else:
-        _MODEL_PROVIDERS_DICT.update({provider_name: {"chat_model": chat_model}})
+        if base_url is not None:
+            _MODEL_PROVIDERS_DICT.update(
+                {provider_name: {"chat_model": chat_model, "base_url": base_url}}
+            )
+        else:
+            _MODEL_PROVIDERS_DICT.update({provider_name: {"chat_model": chat_model}})
 
 
 def batch_register_model_provider(
@@ -159,7 +198,7 @@ def batch_register_model_provider(
         providers: List of ChatModelProvider dictionaries, each containing:
             - provider_name: Name of the provider to register
             - chat_model: Either a BaseChatModel class or a string identifier for a supported provider
-            - base_url: Optional base URL for API endpoints(Optional parameter; effective only when `chat_model` is a string and is "openai-compatible".)
+            - base_url: The API address of the model provider (optional, valid for both types of `chat_model`, but mainly used when `chat_model` is a string and is "openai-compatible")
             - provider_config: The configuration of the model provider(Optional parameter; effective only when `chat_model` is a string and is "openai-compatible".)
                 It can be configured to configure some related parameters of the provider, such as whether to support json_mode structured output mode, the list of supported tool_choice
 

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/src/langchain_dev_utils/embeddings/base.py

@@ -1,7 +1,8 @@
-import os
 from typing import Any, Literal, NotRequired, Optional, TypedDict, Union
 
 from langchain.embeddings.base import Embeddings, _SUPPORTED_PROVIDERS, init_embeddings
+from langchain_core.utils import from_env, secret_from_env
+from pydantic import BaseModel
 
 _EMBEDDINGS_PROVIDERS_DICT = {}
 
@@ -14,6 +15,34 @@ class EmbeddingProvider(TypedDict):
     base_url: NotRequired[str]
 
 
+def _get_base_url_field_name(model_cls: type[BaseModel]) -> str | None:
+    """
+    Return 'base_url' if the model has a field named or aliased as 'base_url',
+    else return 'api_base' if it has a field named or aliased as 'api_base',
+    else return None.
+    The return value is always either 'base_url', 'api_base', or None.
+    """
+    model_fields = model_cls.model_fields
+
+    # try model_fields first
+    if "base_url" in model_fields:
+        return "base_url"
+
+    if "api_base" in model_fields:
+        return "api_base"
+
+    # then try aliases
+    for field_info in model_fields.values():
+        if field_info.alias == "base_url":
+            return "base_url"
+
+    for field_info in model_fields.values():
+        if field_info.alias == "api_base":
+            return "api_base"
+
+    return None
+
+
 def _parse_model_string(model_name: str) -> tuple[str, str]:
     """Parse model string into provider and model name.
 
@@ -56,7 +85,7 @@ def register_embeddings_provider(
     Args:
         provider_name: Name of the provider to register
         embeddings_model: Either an Embeddings class or a string identifier for a supported provider
-        base_url: Optional base URL for API endpoints (required when embeddings_model is a string)
+        base_url: The API address of the Embedding model provider (optional, valid for both types of `embeddings_model`, but mainly used when `embeddings_model` is a string and is "openai-compatible")
 
     Raises:
         ValueError: If base_url is not provided when embeddings_model is a string
@@ -77,8 +106,9 @@ def register_embeddings_provider(
         >>> embeddings = load_embeddings("vllm:qwen3-embedding-4b")
         >>> embeddings.embed_query("hello world")
     """
+
+    base_url = base_url or from_env(f"{provider_name.upper()}_API_BASE", default=None)()
     if isinstance(embeddings_model, str):
-        base_url = base_url or os.getenv(f"{provider_name.upper()}_API_BASE")
         if base_url is None:
             raise ValueError(
                 f"base_url must be provided or set {provider_name.upper()}_API_BASE environment variable when embeddings_model is a string"
@@ -98,9 +128,19 @@ def register_embeddings_provider(
             }
         )
     else:
-        _EMBEDDINGS_PROVIDERS_DICT.update(
-            {provider_name: {"embeddings_model": embeddings_model}}
-        )
+        if base_url is not None:
+            _EMBEDDINGS_PROVIDERS_DICT.update(
+                {
+                    provider_name: {
+                        "embeddings_model": embeddings_model,
+                        "base_url": base_url,
+                    }
+                }
+            )
+        else:
+            _EMBEDDINGS_PROVIDERS_DICT.update(
+                {provider_name: {"embeddings_model": embeddings_model}}
+            )
 
 
 def batch_register_embeddings_provider(
@@ -115,7 +155,7 @@ def batch_register_embeddings_provider(
         providers: List of EmbeddingProvider dictionaries, each containing:
             - provider_name: str - Provider name
             - embeddings_model: Union[Type[Embeddings], str] - Model class or provider string
-            - base_url: Optional[str] - Base URL for API endpoints
+            - base_url: The API address of the Embedding model provider (optional, valid for both types of `embeddings_model`, but mainly used when `embeddings_model` is a string and is "openai-compatible")
 
     Raises:
         ValueError: If any of the providers are invalid
@@ -186,7 +226,7 @@ def load_embeddings(
     embeddings = _EMBEDDINGS_PROVIDERS_DICT[provider]["embeddings_model"]
     if isinstance(embeddings, str):
         if not (api_key := kwargs.get("api_key")):
-            api_key = os.getenv(f"{provider.upper()}_API_KEY")
+            api_key = secret_from_env(f"{provider.upper()}_API_KEY", default=None)()
             if not api_key:
                 raise ValueError(
                     f"API key for {provider} not found. Please set it in the environment."
@@ -203,4 +243,8 @@ def load_embeddings(
             **kwargs,
         )
     else:
+        if base_url := _EMBEDDINGS_PROVIDERS_DICT[provider].get("base_url"):
+            url_key = _get_base_url_field_name(embeddings)
+            if url_key is not None:
+                kwargs.update({url_key: base_url})
         return embeddings(model=model, **kwargs)

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/tests/test_agent.py

@@ -1,6 +1,6 @@
 from dotenv import load_dotenv
 from langchain.agents.structured_output import ToolStrategy
-from langchain_community.chat_models import ChatTongyi
+from langchain_qwq import ChatQwen
 from langchain_core.messages import HumanMessage
 from pydantic import BaseModel, Field
 import pytest
@@ -14,7 +14,7 @@ batch_register_model_provider(
     [
         {
             "provider_name": "dashscope",
-            "chat_model": ChatTongyi,
+            "chat_model": ChatQwen,
         },
         {"provider_name": "zai", "chat_model": "openai-compatible"},
     ]
@@ -33,6 +33,7 @@ def test_prebuilt_agent():
     response = agent.invoke(
         {"messages": [HumanMessage("What's the weather in New York?")]}
     )
+
     assert len(response["messages"]) == 4
 
     assert response["messages"][1].tool_calls[0]["name"] == "get_current_weather"

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/tests/test_load_model.py

@@ -1,6 +1,6 @@
 from dotenv import load_dotenv
-from langchain_community.chat_models import ChatTongyi
 from langchain_core.language_models import BaseChatModel
+from langchain_qwq import ChatQwen
 import pytest
 
 from langchain_dev_utils.chat_models import (
@@ -14,7 +14,7 @@ batch_register_model_provider(
     [
         {
             "provider_name": "dashscope",
-            "chat_model": ChatTongyi,
+            "chat_model": ChatQwen,
         },
         {"provider_name": "zai", "chat_model": "openai-compatible"},
     ]

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/tests/test_model_tool_emulator.py

@@ -1,15 +1,17 @@
 from langchain.tools import tool
-from langchain_community.chat_models import ChatTongyi
+from langchain_qwq import ChatQwen
 from langchain_core.messages import HumanMessage, ToolMessage
 import pytest
 
 from langchain_dev_utils.agents import create_agent
 from langchain_dev_utils.agents.middleware import LLMToolEmulator
 from langchain_dev_utils.chat_models import register_model_provider
+from dotenv import load_dotenv
 
+load_dotenv()
 register_model_provider(
     "dashscope",
-    ChatTongyi,
+    ChatQwen,
 )
 
 

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/tests/test_plan_middleware.py

@@ -54,7 +54,10 @@ def test_plan_middleware():
 
     assert result["plan"]
     assert len(result["plan"]) == 3
-    assert all([plan["status"] == "done" for plan in result["plan"]])
+    assert all([plan["status"] == "done" for plan in result["plan"]]) or all(
+        [plan["status"] == "done" for plan in result["plan"][:-1]]
+        + [result["plan"][-1]["status"] == "in_progress"]
+    )
 
     write_plan_count = 0
     finish_sub_plan_count = 0

{langchain_dev_utils-1.1.12 → langchain_dev_utils-1.1.14}/tests/test_router_model.py

@@ -1,6 +1,6 @@
 from dotenv import load_dotenv
 from langchain.tools import tool
-from langchain_community.chat_models import ChatTongyi
+from langchain_qwq import ChatQwen
 from langchain_core.messages import HumanMessage, ToolMessage
 
 from langchain_dev_utils.agents import create_agent
@@ -13,7 +13,7 @@ batch_register_model_provider(
     [
         {
             "provider_name": "dashscope",
-            "chat_model": ChatTongyi,
+            "chat_model": ChatQwen,
         },
         {"provider_name": "zai", "chat_model": "openai-compatible"},
     ]