[Bug]: Upgrade to version 0.19.1, and an error was reported when asking questions in the chat session.

[doomer2012]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (Language Policy).
• Non-english title submitions will be closed directly ( 非英文标题的提交将会被直接关闭 ) (Language Policy).
• Please do not modify this template :) and fill in all the required fields.

RAGFlow workspace code commit ID

#8187

RAGFlow image version

v0.19.1

Other environment information

ragflow从0.19.0 升级到 0.19.1
服务正常启动后，所有聊天会话，提问都报错，服务端错误日志：
2025-06-23 12:06:42,456 ERROR    14 Base.init() got multiple values for argument 'base_url'
Traceback (most recent call last):
  File "/ragflow/.venv/lib/python3.10/site-packages/flask/app.py", line 880, in full_dispatch_request
    rv = self.dispatch_request()
  File "/ragflow/.venv/lib/python3.10/site-packages/flask/app.py", line 865, in dispatch_request
    return self.ensure_sync(self.view_functions[rule.endpoint])(view_args)  # type: ignore[no-any-return]
  File "/ragflow/.venv/lib/python3.10/site-packages/flask_login/utils.py", line 290, in decorated_view
    return current_app.ensure_sync(func)(*args, kwargs)
  File "/ragflow/api/utils/api_utils.py", line 172, in decorated_function
    return func(*_args, _kwargs)
  File "/ragflow/api/apps/llm_app.py", line 222, in add_llm
    mdl = ChatModel[factory](
  File "/ragflow/rag/llm/chat_model.py", line 459, in init
    super().init(key, model_name, kwargs["base_url"], kwargs)
TypeError: Base.init() got multiple values for argument 'base_url'

Actual behavior

聊天会话提问均报错：

添加模型，提交也报错：

Expected behavior

No response

Steps to reproduce

ragflow从0.19.0 正常升级到 0.19.1。
聊天会话提问，问题复现。

Additional information

No response

[dosubot]

Hi @doomer2012! I'm Dosu and I’m helping the ragflow team.

The error is caused by the way the ChatModel (specifically, classes like AzureChat) initializes its parent Base class: it passes the base_url argument both positionally and inside **kwargs, which leads to Python raising a TypeError about multiple values for the same argument. This was likely introduced or exposed by changes in v0.19.1. You can confirm this in the code, where you’ll see lines like:

super().__init__(key, model_name, kwargs["base_url"], **kwargs)

If kwargs contains a 'base_url' key, this results in two values being passed for base_url. To work around this, you can remove 'base_url' from kwargs before calling super().init, for example:

base_url = kwargs.pop("base_url", None)
super().__init__(key, model_name, base_url, **kwargs)

There are no migration notes or documentation updates for v0.19.1 that mention this change or provide guidance on model initialization or argument handling. No existing issues or pull requests directly address this bug either.
You can safely update your code as above to resolve the error until an official fix is released.
Reference

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}   

[asiroliu]

import from #8187

[shiyang1026]

@KevinHuSh
The same error occurs when I try to create the dataset.
But if I close the graphrag first, the dataset creation works fine without any errors.

Docker logs:

[asiroliu]

@shiyang1026

1. Constructing GraphRAG requires using a chat LLM.
2. Are you also using AzureChat as your LLM?

[liuyao02]

I have used the latest version of Docker after merging #8426, v0.19.1, but I still encounter errors when configuring azure's gpt-4.1:
ERROR: Base.init() got multiple values for argument 'base_url'