oumi-ai/oumi

Easily fine-tune (SFT/RL), evaluate, and deploy Qwen, Gemma, or any open agentic LLM/VLM!

About the project

Oumi is a toolkit for building foundation models, including preparing data, fine-tuning models, evaluating them, and deploying inference endpoints. Researchers and developers use it with open-weight language and vision-language models such as Qwen and Gemma. The catalogue instructions support workflows around operating Oumi.

Latest release v0.9 · 16 Sept 2026

2 files for Codex, OpenCode and Claude Code: oumi AGENTS.md, oumi CLAUDE.md — 508 tokens loaded in every session.

9,425Stars on the repository
2Files it configures its agents with
508Tokens loaded in every session
3Agents configured

Instructions

These files are oumi-ai/oumi's own configuration — they tell Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.