This project provides a proxy server that enables the use of Ollama models as a backend for GitHub Copilot-like code completion in IDEs. It targets developers seeking to leverage local, open-source LLMs for coding assistance, offering a cost-effective and privacy-preserving alternative to proprietary solutions.

How It Works

The proxy acts as an intermediary, intercepting requests from IDE plugins (like copilot.vim, copilot.el, or VS Code's Copilot extension) and forwarding them to a running Ollama instance. It translates the Copilot API format to Ollama's expected input, allowing local models such as codellama:code to provide code suggestions. This approach democratizes AI-powered coding by utilizing accessible local infrastructure.

Quick Start & Requirements

• Install Ollama: curl -fsSL https://ollama.com/install.sh | sh
• Pull a compatible model: ollama pull codellama:code
• Install the proxy: go install github.com/bernardo-bruning/ollama-copilot@latest
• Ensure $HOME/go/bin is in your $PATH.
• Run the proxy: ollama-copilot or OLLAMA_HOST="<ollama_host>" ollama-copilot
• Configure IDE plugins with the proxy address (e.g., http://localhost:11435 or 11437 for VS Code).

Highlighted Details

• Supports copilot.vim, copilot.el (experimental), and VS Code Copilot extension.
• Allows configuration of the Ollama host for remote or containerized deployments.
• Enables local, privacy-focused code completion using open-source LLMs.

Maintenance & Community

The project is maintained by Bernardo Bruning. Further community engagement channels are not explicitly listed in the README.

Licensing & Compatibility

The repository does not explicitly state a license. Compatibility for commercial use or linking with closed-source projects is not specified.

Limitations & Caveats

The project is experimental for Emacs integration. Windows setup and comprehensive documentation are listed as future roadmap items, indicating potential setup hurdles for users on that platform.