Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Ability to use Python LLM package plugins #51

Open
castedo opened this issue Sep 6, 2024 · 2 comments
Open

Ability to use Python LLM package plugins #51

castedo opened this issue Sep 6, 2024 · 2 comments
Assignees

Comments

@castedo
Copy link

castedo commented Sep 6, 2024

This is a feature idea which I might add to copyaid.it. Might be something manubot-ai-editor wants to do too.

The llm Python package has a large number of plugins that can be used to easily (I assume) switch between different LLM APIs and local LLM models:

https://llm.datasette.io/en/stable/plugins/directory.html#plugin-directory

I imagine this plugin API interface is simple enough that it's a quick way to expand out into many different LLM implementations.

@miltondp
Copy link
Collaborator

miltondp commented Sep 6, 2024

Thank you, @castedo. We are currently implementing this feature but with LangChain. We haven't compared it with this package. Do you have any thoughts about this?

@castedo
Copy link
Author

castedo commented Sep 6, 2024

LangChain does much more than just make it easy to switch between LLM sources (i.e., OpenAI vs Anthropic vs local). Whereas calling the llm package plugins via whatever internal interface is used by llm is just one possible easy solution.

If you are using LangChain anyway for other reasons, then ease of LLM source switching is a nice secondary benefit that pops out and there is little reason to use the llm package interface.

But if all you want to achieve is ease of LLM source switching then LangChain is way overkill and will pull in a ton of dependencies (I think). llm is just a simple CLI tool kind of similar copyaid and manubot-ui-editor in CLI form.

@falquaddoomi falquaddoomi self-assigned this Sep 25, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants