Ronin

Native intelligence for Linux

Threaded AI Chat

Organise conversations in persistent threads. Switch models mid-flow without losing context.

Memory & Artifacts

Local-first storage for memories and reusable artifacts. Attach context with a single @ mention.

Hybrid Model Support

Run local models via Ollama or connect to cloud providers. Per-provider settings let you tailor each model.

Fast CLI Launcher

Summon Ronin from the terminal. Native performance, zero Electron overhead.

Built with

Rust GPUI SQLite Tokio Ollama OpenRouter