GoogleSerperResults._run() got an unexpected keyword argument '__arg1' while using bind_tools

[akashDawariLTIM]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain.tools.google_serper.tool import GoogleSerperResults
from langchain_openai import AzureChatOpenAI
import os

lm = AzureChatOpenAI(model=os.getenv("DEPLOYMENT_NAME"),
                            deployment_name=os.getenv("DEPLOYMENT_NAME"),
                            api_key=os.getenv("API_KEY"),
                            api_version=os.getenv("API_VERSION"),
                            azure_endpoint=os.getenv("API_BASE"),
                            temperature=0)


os.environ["SERPER_API_KEY"] = os.getenv("SERPER_API_KEY")

serper = GoogleSerperResults()
tools = [serper]


# Modification: tell the LLM which tools it can call
llm_with_tools = llm.bind_tools(tools)

Description

Getting error that GoogleSerperResults._run() got an unexpected keyword argument '__arg1'

can anyone help me with that?

System Info

Python 3.10.9

[dosubot[bot]]

The error GoogleSerperResults._run() got an unexpected keyword argument '__arg1' suggests that the bind_tools method is passing an unexpected keyword argument to the _run method of the GoogleSerperResults class.

The _run method of GoogleSerperResults is defined to accept only query and run_manager as arguments. The error indicates that an additional argument __arg1 is being passed, which is not expected by the method.

To resolve this issue, you can modify the _run method to accept additional keyword arguments:

class GoogleSerperResults(BaseTool):
    # ... existing code ...

    def _run(
        self,
        query: str,
        run_manager: Optional[CallbackManagerForToolRun] = None,
        **kwargs  # Accept additional keyword arguments
    ) -> str:
        """Use the tool."""
        return str(self.api_wrapper.results(query))

This modification allows the _run method to accept additional keyword arguments without raising an error. However, this is a workaround, and ideally, the bind_tools method should be checked to ensure it passes the correct arguments.

Additionally, ensure that the tools you are binding are compatible with the expected format and that the tool_choice parameter is correctly specified if used.

Details

Sources

• GoogleSerperRun._arun() got an unexpected keyword argument 'arg1'
• libs/community/langchain_community/tools/google_serper/tool.py
• libs/partners/openai/langchain_openai/chat_models/base.py
• libs/core/langchain_core/utils/function_calling.py
• libs/partners/groq/langchain_groq/chat_models.py
• libs/partners/anthropic/langchain_anthropic/chat_models.py
• docs/docs/how_to/tool_calling.ipynb
• libs/core/langchain_core/tools.py
• libs/partners/huggingface/tests/unit_tests/test_chat_models.py

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.

[nick-youngblut]

Getting the same for PubmedQueryRun => TypeError: PubmedQueryRun._run() got an unexpected keyword argument '__arg1'

We shouldn't need to modify the tool just to get it running, as described above.