how to solve langgraph or langchain init_chat_model authentication error

Jun 20, 2025

1. Purpose

This post aims to solve the authentication error when using langgraph or langchain init_chat_model.

Traceback (most recent call last):
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/langgraph/basic/stategraph1.py", line 61, in <module>
    stream_graph_updates(user_input)
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/langgraph/basic/stategraph1.py", line 47, in stream_graph_updates
    for event in graph.stream({"messages": [{"role":"user", "content": user_input}]}):
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/langgraph/pregel/__init__.
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/langgraph/basic/stategraph1.py", line 29, in chatbot
    return {"messages": [llm.invoke(state["messages"])]}
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/langchain_core/language_models/chat_models.py", line 371, in invoke
    self.generate_prompt(
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/langchain_core/language_models/chat_models.py", line 956, in generate_prompt
    return self.generate(prompt_messages, stop=stop, callbacks=callbacks, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/langchain_core/language_models/chat_models.py", line 775, in generate
    self._generate_with_cache(
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/langchain_core/language_models/chat_models.py", line 1021, in _generate_with_cache
    result = self._generate(
             ^^^^^^^^^^^^^^^
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/langchain_deepseek/chat_models.py", line 296, in _generate
    return super()._generate(
           ^^^^^^^^^^^^^^^^^^
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/langchain_openai/chat_models/base.py", line 973, in _generate
    response = self.client.create(**payload)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/openai/_utils/_utils.py", line 287, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/openai/resources/chat/completions/completions.py", line 925, in create
    return self._post(
           ^^^^^^^^^^^
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/openai/_base_client.py", line 1239, in post
    return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/bswen/projects/PythonProjects/langchain-rag1/.venv/lib/python3.12/site-packages/openai/_base_client.py", line 1034, in request
    raise self._make_status_error_from_response(err.response) from None
openai.AuthenticationError: Error code: 401 - {'error': {'message': 'Authentication Fails, Your api key: **** is invalid', 'type': 'authentication_error', 'param': None, 'code': 'invalid_request_error'}}
During task with name 'chatbot' and id '82a05b24-8d68-b3c6-7a83-c538aff44494'

2. The code and solution

2.1 The code

Here is the main code for testing langgraph basic graph:

from typing import Annotated
from typing_extensions import TypedDict
from langgraph.graph import StateGraph,START
from langgraph.graph.message import add_messages
from langchain.chat_models import init_chat_model



default_api_key= "mykey....";
default_base_url='https://api.siliconflow.cn/v1'
default_model = "deepseek-ai/DeepSeek-V3"

# the state node
class State(TypedDict):
    messages: Annotated[list,add_messages]

graph_builder = StateGraph(State)

# init the chat model
llm = init_chat_model(model=default_model,
                      api_key=default_api_key, base_url=default_base_url)


def chatbot(state:State):
    return {"messages": [llm.invoke(state["messages"])]}

graph_builder.add_node("chatbot", chatbot)
graph_builder.add_edge(START, "chatbot")

# compile the graph
graph = graph_builder.compile()


def stream_graph_updates(user_input: str):
    for event in graph.stream({"messages": [{"role":"user", "content": user_input}]}):
        for value in event.values():
            print("Assistant:", value["messages"][-1].content)

# start the graph
while True:
    try:
        user_input = input("User: ")
        if user_input.lower() in ["quit", "exit", "q"]:
            print("Goodbye")
            break
        stream_graph_updates(user_input)
    except:
        user_input = "What do you know about LangGraph?"
        print("User: "+user_input)
        stream_graph_updates(user_input)
        break

I want to build a graph like this:

You can see that I have used the init_chat_model function to initialize a chat model as follows:

llm = init_chat_model(model=default_model,
                      model_provider="openai",
                      api_key=default_api_key, base_url=default_base_url)

But when I run the code, I get the following error:

openai.AuthenticationError: Error code: 401 - {'error': {'message': 'Authentication Fails, Your api key: **** is invalid', 'type': 'authentication_error', 'param': None, 'code': 'invalid_request_error'}}
During task with name 'chatbot' and id '82a05b24-8d68-b3c6-7a83-c538aff44494'

2.2 The solution

Let’s view the documentation for init_chat_model:

Here is the source code for init_chat_model function:

def init_chat_model(
    model: Optional[str] = None,
    *,
    model_provider: Optional[str] = None,
    configurable_fields: Optional[
        Union[Literal["any"], list[str], tuple[str, ...]]
    ] = None,
    config_prefix: Optional[str] = None,
    **kwargs: Any,
) -> Union[BaseChatModel, _ConfigurableModel]:
    """Initialize a ChatModel from the model name and provider.

    **Note:** Must have the integration package corresponding to the model provider
    installed.

    Args:
        model: The name of the model, e.g. "o3-mini", "claude-3-5-sonnet-latest". You can
            also specify model and model provider in a single argument using
            '{model_provider}:{model}' format, e.g. "openai:o1".
        model_provider: The model provider if not specified as part of model arg (see
            above). Supported model_provider values and the corresponding integration
            package are:

            - 'openai'              -> langchain-openai
            - 'anthropic'           -> langchain-anthropic
            - 'azure_openai'        -> langchain-openai
            - 'azure_ai'            -> langchain-azure-ai
            - 'google_vertexai'     -> langchain-google-vertexai
            - 'google_genai'        -> langchain-google-genai
            - 'bedrock'             -> langchain-aws
            - 'bedrock_converse'    -> langchain-aws
            - 'cohere'              -> langchain-cohere
            - 'fireworks'           -> langchain-fireworks
            - 'together'            -> langchain-together
            - 'mistralai'           -> langchain-mistralai
            - 'huggingface'         -> langchain-huggingface
            - 'groq'                -> langchain-groq
.....

I am using a openAI compatible model from siliconflow, which is compatible with openAI. And I notice that the init_chat_model has a parameter named model_provider which will impact the model authentication process ,So I need to set model_provider to initialize a chat model as follows:

llm = init_chat_model(model=default_model,
                      model_provider="openai",
                      api_key=default_api_key, base_url=default_base_url)

After setting model_provider to "openai", the code runs successfully.

The key solution is that, you should set the model_provider to "openai", if not, langgraph will use the inferred model provider to authenticate the model, which will be deepseek, and this is not the right model provider.

3. The theory

What’s the difference between langchain and langgraph?

LangGraph is a graph-based language model, which is a new way to use language models. It is based on the idea that language models can be used to build graphs, and then the graphs can be used to generate text.

But langchain is a library for building LLM applications. It is a high-level library that provides a set of tools for building LLM applications. It is not a graph-based language model.

Here is a diagram shows the difference:

4. Summary

In this post, I have explained how to use langgraph to build a graph-based language model, and how to solve the authentication problem when using init_chat_model to initialize a chat model. You should pay attention to the model_provider parameter when using init_chat_model to initialize a chat model if you encounter authentication problems.

Final Words + More Resources

My intention with this article was to help others who might be considering solving such a problem. So I hope that’s been the case here. If you still have any questions, don’t hesitate to ask me by email: Email me

Here are also the most important links from this article along with some further resources that will help you in this scope:

👨‍💻 init_chat_model api
👨‍💻 langgraph

Oh, and if you found these resources useful, don’t forget to support me by starring the repo on GitHub!