huggingface / llm-ls

LSP server leveraging LLMs for code completion (and more?)
Apache License 2.0
602 stars 49 forks source link

Breaking change: Support OpenAI API, arbitrary APIs #42

Closed rhelmot closed 9 months ago

rhelmot commented 11 months ago

With this change and some minor tweaks to llm.nvim, I was able to get code completions from a locally running copy of CodeLlama-13b through an OpenAI-compatible frontend.

McPatate commented 11 months ago

Hello @rhelmot, thank you for your contribution. What OpenAI-compatible frontend are we talking about here?

I took a quick glance, it looks good. I'll take a deeper look when I have a bit more time.

rhelmot commented 11 months ago

I was using python -m llama_cpp.server, but I believe there are more as well. litellm comes to mind.

McPatate commented 9 months ago

Closing in favor of #40