# Oumi

> Oumi (Open Universal Machine Intelligence) is an open-source, Apache-2.0 platform for ML engineers and researchers to train, fine-tune, evaluate, and deploy foundation models (LLMs and VLMs) through one unified interface. It ships as a Python library and a first-party CLI, with a hosted managed platform and an MCP server for agentic clients. Generated by the API Evangelist enrichment pipeline from the public Oumi surface.

## Product
- [Website](https://oumi.ai): Oumi AI model development platform.
- [Managed Platform](https://platform.oumi.ai): Hosted Free/Pro/Enterprise product; sign up and run jobs.
- [Pricing](https://oumi.ai/pricing): Free ($0, up to $50 credits), Pro ($25/mo + pay-as-you-go), Enterprise (custom).

## Docs
- [Documentation](https://oumi.ai/docs): Open-source platform docs for training, inference, evaluation, deployment.
- [Quickstart](https://oumi.ai/docs/en/latest/get_started/quickstart.html): Get started training a model from a recipe config.
- [API Reference](https://oumi.ai/docs/en/latest/api/oumi.html): Python module reference (oumi.core, oumi.datasets, oumi.inference, ...).
- [CLI Reference](https://oumi.ai/docs/en/latest/cli/commands.html): The `oumi` command surface.

## Tooling
- [Python Package (PyPI)](https://pypi.org/project/oumi/): `pip install oumi` — library + `oumi` CLI (train, evaluate, infer, launch, deploy, analyze, synth, tune).
- [oumi-mcp](https://github.com/oumi-ai/oumi): First-party Model Context Protocol server for Claude/Cursor (v0.8+).

## Source & Community
- [GitHub](https://github.com/oumi-ai/oumi): Apache-2.0 source repository (9.4k+ stars).
- [Releases / Changelog](https://github.com/oumi-ai/oumi/releases): Semver releases (current 0.8).
- [Discord](https://discord.gg/oumi): Community support.
- [Blog / Newsletter](https://oumiai.substack.com): Oumi AI Substack.