hugging face basic llm endpoint is implemented

[huzaifansari54]

Feature #19
Description
This pull request is add support for hugging face llms in lang chain

Dependencies
Updated pubspec.yaml to reflect new dependencies
Added a Huggingface_client

Testing
use an example for test the model

[huzaifansari54]

documentation is left i am busy that why i cant add chat model as well let me know if there is any changes required

[davidmigloz]

Hey @huzaifansari54,

Thanks for your work!
I'll review it soon (can't promise when, I'm a bit busy at the moment)

[huzaifansari54]

stream api is add for llms

[davidmigloz]

awesome, thanks! I'll try to start the review today

[hajajmaor]

Update on this PR? @huzaifansari54

[huzaifansari54]

It is under review now

[hajajmaor]

Tried to use it to create agent, got error:

The argument type 'HuggingfaceInference' can't be assigned to the parameter type 'BaseChatModel'. dartargument_type_not_assignable

final llm = HuggingfaceInference.call(
      apiKey: Env.huggingFaceApiKey,
      model: modelName!,
    );
    final tools = [tool];
final agent = ToolsAgent.fromLLMAndTools(
      llm: llm,
      tools: tools,
    );

Hope to see this PR fixed and merged soon

[davidmigloz]

going back to this PR soon! sorry for the delay

[hajajmaor]

Hey, any updates?

[davidmigloz]

@gemini-cli /review

[github-actions]

📋 Review Summary

This PR adds support for Hugging Face Inference API. The implementation looks good, but there are a few areas for improvement.

🔍 General Feedback

• The main points are handling of nullable responses from the API client to avoid potential runtime errors.
• Completing the copyWith method in HuggingFaceOptions.
• Improving the example by not hardcoding the API key.

Overall, a good contribution that extends the functionality of LangChain.dart.