### Everything you need to build state-of-the-art foundation models, end-to-end
## News
- [2026/08] Extended GRPO reinforcement learning to support tool use
- [2026/07] Added support for tools, environments (simulated, lookup, database), and agentic data synthesis
- [2026/06] Added support for the Gemma 4 model family
- [2026/06] Added partial-failure support across inference, judging, and data synthesis
- [2026/05] [Oumi v0.8 released](https://github.com/oumi-ai/oumi/releases/tag/v0.8) with `oumi deploy` CLI for dedicated inference endpoints, an `oumi-mcp` MCP server for Claude/Cursor integration, batch API support across Anthropic/Fireworks/Together, and Transformers v5 / TRL / vLLM dependency upgrades
- [2026/03] Upgraded to Transformers v5, TRL v0.30, vLLM v0.19, and veRL v0.7 compatibility
- [2026/03] [MCP Integration Phase 1](https://github.com/oumi-ai/oumi/pull/2234): package scaffold and dependencies for MCP server support
- [2026/03] New: `oumi deploy` command for deploying oumi models dedicated inference endpoints on fireworks.ai and parasail
- [2026/03] Added support for Qwen3.5 model family
- [2026/03] Inference engines received multiple improvements: list_models api, improved error reporting
- [2026/02] [Preview of using the Oumi Platform and Lambda to fine-tune and deploy a 4B model for user intent classification](https://youtu.be/0XpfYRpd_FA)
- [2026/02] [Lambda and Oumi partner for end-to-end custom model development](https://blog.oumi.ai/p/lambda-and-oumi-partner-for-end-to)
- [2025/12] [Oumi v0.6.0 released](https://github.com/oumi-ai/oumi/releases/tag/v0.6.0) with Python 3.13 support, `oumi analyze` CLI command, TRL 0.26+ support, and more
- [2025/12] [WeMakeDevs AI Agents Assemble Hackathon: Oumi webinar on Finetuning for Text-to-SQL](https://www.youtube.com/watch?v=6wPikqRZ7bQ&t=3203s)
- [2025/12] [Oumi co-sponsors WeMakeDevs AI Agents Assemble Hackathon with over 2000 project submissions](https://www.wemakedevs.org/hackathons/assemblehack25)
- [2025/11] [Oumi v0.5.0 released](https://github.com/oumi-ai/oumi/releases/tag/v0.5) with advanced data synthesis, hyperparameter tuning automation, support for OpenEnv, and more
Older updates
- [2025/11] [Example notebook to perform RLVF fine-tuning with OpenEnv](https://github.com/oumi-ai/oumi/blob/main/notebooks/Oumi%20-%20OpenEnv%20GRPO%20with%20trl.ipynb), an open source library from the Meta PyTorch team for creating, deploying, and distributing agentic RL environments
- [2025/10] [Oumi v0.4.1](https://github.com/oumi-ai/oumi/releases/tag/v0.4.1) and [v0.4.2](https://github.com/oumi-ai/oumi/releases/tag/v0.4.2) released] with support for Qwen3-VL and Transformers v4.56, data synthesis documentation and examples, and many bug fixes
- [2025/09] [Oumi v0.4.0 released](https://github.com/oumi-ai/oumi/releases/tag/v0.4.0) with DeepSpeed support, a Hugging Face Hub cache management tool, KTO/Vision DPO trainer support
- [2025/08] Training and inference support for OpenAI's `gpt-oss-20b` and `gpt-oss-120b`: [recipes here](https://github.com/oumi-ai/oumi/tree/main/configs/recipes/gpt_oss)
- [2025/08] Aug 14 Webinar - [OpenAI's gpt-oss: Separating the Substance from the Hype](https://youtu.be/g1PkAV7fXn0).
- [2025/08] [Oumi v0.3.0 released](https://github.com/oumi-ai/oumi/releases/tag/v0.3.0) with model quantization (AWQ), an improved LLM-as-a-Judge API, and Adaptive Inference
- [2025/07] Recipe for [Qwen3 235B](https://github.com/oumi-ai/oumi/blob/main/configs/recipes/qwen3/inference/235b_a22b_together_infer.yaml)
- [2025/07] July 24 webinar: ["Training a State-of-the-art Agent LLM with Oumi + Lambda"](https://youtu.be/f3SU_heBP54)
- [2025/06] [Oumi v0.2.0 released](https://github.com/oumi-ai/oumi/releases/tag/v0.2.0) with support for GRPO fine-tuning, a plethora of new model support, and much more
- [2025/06] Announcement of [Data Curation for Vision Language Models (DCVLR) competition](https://oumi.ai/blog/posts/announcing-dcvlr) at NeurIPS2025
- [2025/06] Recipes for training, inference, and eval with the newly released [Falcon-H1](https://github.com/oumi-ai/oumi/tree/main/configs/recipes/falcon_h1) and [Falcon-E](https://github.com/oumi-ai/oumi/tree/main/configs/recipes/falcon_e) models
- [2025/05] Support and recipes for [InternVL3 1B](https://github.com/oumi-ai/oumi/tree/main/configs/recipes/vision/internvl3)
- [2025/04] Added support for training and inference with Llama 4 models: Scout (17B activated, 109B total) and Maverick (17B activated, 400B total) variants, including full fine-tuning, LoRA, and QLoRA configurations
- [2025/04] Recipes for [Qwen3 model family](https://github.com/oumi-ai/oumi/tree/main/configs/recipes/qwen3)
- [2025/04] Introducing HallOumi: a State-of-the-Art Claim-Verification Model [(technical overview)](https://oumi.ai/blog/posts/introducing-halloumi)
- [2025/04] Oumi now supports two new Vision-Language models: [Phi4](https://github.com/oumi-ai/oumi/tree/main/configs/recipes/vision/phi4) and [Qwen 2.5](https://github.com/oumi-ai/oumi/tree/main/configs/recipes/vision/qwen2_5_vl_3b)
## About
Oumi is a fully open-source platform that streamlines the entire lifecycle of foundation models - from data preparation and training to evaluation and deployment. Whether you're developing on a laptop, launching large scale experiments on a cluster, or deploying models in production, Oumi provides the tools and workflows you need.
With Oumi, you can:
- Train and fine-tune models from 10M to 405B parameters using state-of-the-art techniques (SFT, LoRA, QLoRA, GRPO, and more)
- Work with both text and multimodal models (Llama, DeepSeek, Qwen, Phi, and others)
- Synthesize and curate training data with LLM judges
- ⚡️ Deploy models efficiently with popular inference engines (vLLM, SGLang)
- Evaluate models comprehensively across standard benchmarks
- Run anywhere - from laptops to clusters to clouds (AWS, Azure, GCP, Lambda, and more)
- Integrate with both open models and commercial APIs (OpenAI, Anthropic, Vertex AI, Together, Parasail, ...)
All with one consistent API, production-grade reliability, and all the flexibility you need for research.
Learn more at [oumi.ai](https://oumi.ai/docs), or jump right in with the [quickstart guide](https://oumi.ai/docs/en/latest/get_started/quickstart.html).
## Getting Started
| **Notebook** | **Try in Colab** | **Goal** |
|----------|--------------|-------------|
| ** Getting Started: A Tour** |
| Quick tour of core features: training, evaluation, inference, and job management |
| ** Model Finetuning Guide** |
| End-to-end guide to LoRA tuning with data prep, training, and evaluation |
| ** Model Distillation** |
| Guide to distilling large models into smaller, efficient ones |
| ** Model Evaluation** |
| Comprehensive model evaluation using Oumi's evaluation framework |
| **☁️ Remote Training** |
| Launch and monitor training jobs on cloud (AWS, Azure, GCP, Lambda, etc.) platforms |
| ** LLM-as-a-Judge** |
| Filter and curate training data with built-in judges |
## Usage
### Installation
Choose the installation method that works best for you:
Using pip (Recommended)
```bash
# Basic installation
uv pip install oumi
# With GPU support
uv pip install 'oumi[gpu]'
# Latest development version
uv pip install git+https://github.com/oumi-ai/oumi.git
```
Don't have uv? [Install it](https://docs.astral.sh/uv/getting-started/installation/) or use `pip` instead.
Using Docker
```bash
# Pull the latest image
docker pull ghcr.io/oumi-ai/oumi:latest
# Run oumi commands
docker run --gpus all -it ghcr.io/oumi-ai/oumi:latest oumi --help
# Train with a mounted config
docker run --gpus all -v $(pwd):/workspace -it ghcr.io/oumi-ai/oumi:latest \
oumi train --config /workspace/my_config.yaml
```
Quick Install Script (Experimental)
Try Oumi without setting up a Python environment. This installs Oumi in an isolated environment:
```bash
curl -LsSf https://oumi.ai/install.sh | bash
```
For more advanced installation options, see the [installation guide](https://oumi.ai/docs/en/latest/get_started/installation.html).
### Oumi CLI
You can quickly use the `oumi` command to train, evaluate, and infer models using one of the existing [recipes](/configs/recipes):
```shell
# Training
oumi train -c configs/recipes/smollm/sft/135m/quickstart_train.yaml
# Evaluation
oumi evaluate -c configs/recipes/smollm/evaluation/135m/quickstart_eval.yaml
# Inference
oumi infer -c configs/recipes/smollm/inference/135m_infer.yaml --interactive
```
For more advanced options, see the [training](https://oumi.ai/docs/en/latest/user_guides/train/train.html), [evaluation](https://oumi.ai/docs/en/latest/user_guides/evaluate/evaluate.html), [inference](https://oumi.ai/docs/en/latest/user_guides/infer/infer.html), and [llm-as-a-judge](https://oumi.ai/docs/en/latest/user_guides/judge/judge.html) guides.
### Running Jobs Remotely
You can run jobs remotely on cloud platforms (AWS, Azure, GCP, Lambda, etc.) using the `oumi launch` command:
```shell
# GCP
oumi launch up -c configs/recipes/smollm/sft/135m/quickstart_gcp_job.yaml
# AWS
oumi launch up -c configs/recipes/smollm/sft/135m/quickstart_gcp_job.yaml --resources.cloud aws
# Azure
oumi launch up -c configs/recipes/smollm/sft/135m/quickstart_gcp_job.yaml --resources.cloud azure
# Lambda
oumi launch up -c configs/recipes/smollm/sft/135m/quickstart_gcp_job.yaml --resources.cloud lambda
```
**Note:** Oumi is in beta and under active development. The core features are stable, but some advanced features might change as the platform improves.
## Why use Oumi?
If you need a comprehensive platform for training, evaluating, or deploying models, Oumi is a great choice.
Here are some of the key features that make Oumi stand out:
- **Zero Boilerplate**: Get started in minutes with ready-to-use recipes for popular models and workflows. No need to write training loops or data pipelines.
- **Enterprise-Grade**: Built and validated by teams training models at scale
- **Research Ready**: Perfect for ML research with easily reproducible experiments, and flexible interfaces for customizing each component.
- **Broad Model Support**: Works with most popular model architectures - from tiny models to the largest ones, text-only to multimodal.
- **SOTA Performance**: Native support for distributed training techniques (FSDP, DeepSpeed, DDP) and optimized inference engines (vLLM, SGLang).
- **Community First**: 100% open source with an active community. No vendor lock-in, no strings attached.
## Examples & Recipes
Explore the growing collection of ready-to-use configurations for state-of-the-art models and training workflows:
**Note:** These configurations are not an exhaustive list of what's supported, simply examples to get you started. You can find a more exhaustive list of supported [models](