LLM powered development for Neovim
A Neovim plugin, formerly hfcc.nvim, that adds Copilot-style ghost-text code completion backed by the Hugging Face Inference API, Ollama, or another HTTP model endpoint.
llm.nvim is a Neovim plugin for LLM-assisted development, using a separate backend called llm-ls. It adds ghost-text code completion similar to Copilot, sending an HTTP request built from the surrounding code to a configured model.
It can use the Hugging Face Inference API, an Ollama server, or any other HTTP endpoint that follows one of its supported backend APIs, with the request always sized to fit the model's context window using tokenizers to count tokens. The model, backend URL and API token can each be set through plugin options or environment variables, and the plugin can automatically complete a backend's URL with the right path unless that behavior is disabled.
Free use of the Hugging Face Inference API is rate-limited; a paid PRO plan removes that limit.