The "LLM Commit" extension generates Git commit messages using OpenAI-compatible APIs. It helps you create concise, imperative commit messages based on the staged changes in your repository.
Requirements
Visual Studio Code version 1.80.0 or higher
An OpenAI-compatible API endpoint (such as OpenAI, LM Studio, Ollama with OpenAI compatibility, or other compatible services)
Installation
Install Visual Studio Code
Install the "LLM Commit" extension from the VS Code marketplace.
Configure your provider in the extension settings (LM Studio is the default).
Quick Setup
LM Studio (default): Start LM Studio, load a model, and enable the local server - no configuration needed!
Ollama: Run ollama serve to start the server - no configuration needed!
OpenAI: Set provider to "OpenAI" and add your API key in settings
Custom: Set provider to "Custom" and configure URL, API key, and model manually
Model choice
Any instruction following model can accomplish this task, I personally recommend qwen2.5-coder-7b-instruct at the moment since it's a small, fast model that understands code well.
Extension Settings
This extension contributes the following settings:
Provider Presets
llm-commit.provider: Select a provider preset for quick setup:
LM Studio (default): http://127.0.0.1:1234/v1/chat/completions