Repolode
Submit
0xPlaygrounds

0xPlaygrounds/rig

A Rust library for building LLM applications, plus a terminal coding agent built on it whose features are Bevy plugins.

View on GitHubWebsite
Stars
8,926
Forks
991
Open issues
76
Last commit
13 hours ago

What is it?

Rig is a Rust library for building LLM applications with a provider-neutral interface. It offers one API across more than 20 model providers and more than 10 vector stores, covering completion, embedding, transcription, audio and image generation. Agents support tools, multi-turn streaming and typed hooks.

The project also ships `rig`, a terminal coding agent built on the library. The agent is a Bevy app whose tools, commands, sessions, sign-in and terminal view are each a plugin listed in a config file, so parts can be removed, replaced or added. `/reload` rebuilds the agent with plugin changes while keeping the previous build running if the new one fails to compile. It can also be driven from scripts via JSON over pipes.

The workspace is split into companion crates for memory, effect recording and replay, coding tools, and integrations with databases such as PostgreSQL, Qdrant and MongoDB, along with cloud providers like AWS Bedrock and Google Vertex AI. The core and agent runtime also compile to browser WASM. The project is pre-1.0, so releases may include breaking changes.

Key features

  • 01Unified interface across 20+ LLM providers and 10+ vector stores
  • 02Agents with tool calling, multi-turn streaming and typed hooks
  • 03Completion, embedding, transcription, audio and image generation support
  • 04Terminal coding agent whose tools and views are Bevy plugins
  • 05Hot rebuild with `/reload` that keeps the old build if compilation fails
  • 06Browser WASM target for the core and agent runtime

Use cases

  • Building Rust backends that call LLMs and switch providers without rewriting code
  • Creating RAG pipelines backed by PostgreSQL, Qdrant or other vector stores
  • Running a scriptable terminal coding agent that can be extended with plugins
  • Embedding agent logic in browser applications via WASM

Quick start

  1. 1

    Install the coding agent with Cargo

    cargo install rig
  2. 2

    Set a provider API key (such as OPENAI_API_KEY) or sign in with /login chatgpt inside the agent

  3. 3

    Start the coding agent in the terminal

    rig

Things to know

  • Rig is pre-1.0, so releases contain breaking changes.
  • The coding agent needs Rust 1.97.1 or newer and runs only on Linux and macOS.
  • Using the agent requires an API key from a model provider or a ChatGPT plan.

Topics

#llm#rust#agent#ai-agent#vector-database#generative-ai#rag#ai#artificial-intelligence#automation

FAQ

What is rig?
A Rust library for building LLM applications, plus a terminal coding agent built on it whose features are Bevy plugins. Rig is a Rust library for building LLM applications with a provider-neutral interface. It offers one API across more than 20 model providers and more than 10 vector stores, covering completion, embedding, transcription, audio and image generation. Agents support tools, multi-turn streaming and typed hooks.
Is rig free to use? What license does it have?
rig is open source under the MIT license. Check the license terms before using it in commercial products.
Is rig still maintained?
Actively maintained. The last commit to rig was on Oct 20, 2026.
What can I use rig for?
Building Rust backends that call LLMs and switch providers without rewriting code Creating RAG pipelines backed by PostgreSQL, Qdrant or other vector stores Running a scriptable terminal coding agent that can be extended with plugins Embedding agent logic in browser applications via WASM
How do I install rig?
Install the coding agent with Cargo: cargo install rig Set a provider API key (such as OPENAI_API_KEY) or sign in with /login chatgpt inside the agent Start the coding agent in the terminal: rig
Where can I get rig?
The source code and installation instructions are on GitHub: https://github.com/0xPlaygrounds/rig

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects