LocalAI Integration

[malpou]

This draft PR introduces support for integrating LocalAI into our application. Users who wish to run the LocalAI model locally can now do so with ease, leveraging the instructions and settings provided in our readme.

Readme Additions:

Use LocalAI

We have added support to use LocalAI as a model. This is a free and open source model that you can run locally. To use this model, you need to have LocalAI running on your machine.

Setup LocalAI

This is the way our team configured our enviroment to run LocalAI. You can follow these steps here:

1. Follow the LocalAI Getting Started Example
2. Set the OCO_OPENAI_BASE_PATH to the path of your LocalAI instance. For example, if you are running LocalAI on port 8080, you would set the OCO_OPENAI_BASE_PATH to http://localhost:8080/v1.
3. Set the OCO_MODEL to whatever model you're using following the getting started example the value will be ggml-gpt4all-j.
4. Now you can use OpenCommit with LocalAI!

Notes:
I am currently facing some challenges with the LocalAi container, so the full functionality has not been tested exhaustively. I'm trying to resolve the issue and awaiting a response from the team from LocalAi on what could be causing this issue.
Since I've just been following the guide in their documentation. LocalAi Issue

But Local model support should be arriving soon.

Any contributions or feedback on this draft is highly welcome.

Issue #236

[di-sukharev]

@malpou you are superhero!

let me know when you are ready to test this and if any help is needed.

[malpou]

Will do, just waiting atm 😀

[github-actions]

Stale pull request message

[jaroslaw-weber]

No activity? Could i work on itegrating ollama instead? I was able to make it work with another repo.
https://github.com/insulineru/ai-commit/pull/18/files

[intothemoonlite]

ollama is getting very popular! would be good to have

[di-sukharev]

we now have ollama