A desktop AI client that works with any model provider - OpenAI, Anthropic, Google, local models, and everything in between.
Like Codex or Co-Work, but provider-agnostic. Bring your own API keys, pick your model, and work across providers from a single native window - no silos, no vendor lock-in.
Built with Tauri and powered by Inference Gateway.
On first run, the app downloads the infer CLI binary and installs it to ~/.infer/bin/infer. The CLI manages the gateway server and routes requests to whatever provider you configure - OpenAI, Anthropic, Google, local Ollama models, or any OpenAI-compatible endpoint.
The gateway binary lands at ~/.infer/bin/inference-gateway, config lives under ~/.infer/, and agent file tools are scoped to your home directory.
| Platform | Asset name |
|---|---|
| Linux amd64 | infer-linux-amd64 |
| Linux arm64 | infer-linux-arm64 |
| macOS amd64 | infer-darwin-amd64 |
| macOS arm64 | infer-darwin-arm64 |
| Windows amd64 | infer-windows-amd64 |
| Windows arm64 | infer-windows-arm64 |
See CONTRIBUTING.md for prerequisites, local setup, and building.