rig
⚙️🦀 Build modular and scalable LLM Applications in Rust
What is rig?
⚙️🦀 Build modular and scalable LLM Applications in Rust
Overview of rig
What it is
Rig is a Rust library for building LLM applications, and a terminal coding agent built on it that can be extended with Bevy plugins. The library provides provider-neutral interfaces for completion, embedding, transcription, audio, and image generation, along with agents that support tools, multi-turn streaming, and typed hooks. The coding agent is a Bevy app where tools, commands, sessions, sign-in, and the terminal view are each a plugin that can be turned on or off in settings.toml.
Key features
- 20+ model providers and 10+ vector stores, each under one interface
- Agents with tools, multi-turn streaming and typed hooks
- Completion, embedding, transcription, audio and image generation
- GenAI semantic conventions for OpenTelemetry
- Recording and replay of an agent's effects (
rig-cassette) - Browser WASM (
wasm32-unknown-unknown) for the core and the agent runtime - The coding agent rebuilds itself:
/reloadrebuilds the agent with your plugin crates and restarts it in the same session, keeping the old build running if the new one does not compile
Use cases
Building modular and scalable LLM applications in Rust. The library can be used to create agents that use tools, perform multi-turn conversations, and integrate with various model providers and vector stores. The coding agent can be used as a terminal-based assistant that can be extended with custom plugins, and it can be scripted over pipes using rig -p --json or rig --json.
Getting started
To install the coding agent:bashcargo install rig-clirigTo add the library to a project:bashcargo add rigFor more details, see the repository README.
License and pricing
Rig is MIT licensed. It is open source with no stated pricing.
Based on the project's public README and repository metadata.