AI MLOps Framework
9.2k 2026-04-18
oumi-ai/oumi
An end-to-end platform for fine-tuning, evaluating, and deploying open-source Large Language Models (LLMs) and Vision Language Models (VLMs).
Core Features
End-to-end lifecycle management for LLMs/VLMs (fine-tuning, evaluation, deployment)
Support for various open-source models like gpt-oss, Qwen3, DeepSeek-R1
Advanced capabilities including data synthesis, hyperparameter tuning, and model quantization
Deployment to dedicated inference endpoints on commercial providers like fireworks.ai and parasail
CLI tools for model analysis and deployment
Quick Start
pip install oumiDetailed Introduction
Oumi is a comprehensive AI MLOps framework designed to streamline the entire lifecycle of building state-of-the-art foundation models. It empowers developers and researchers to easily fine-tune, evaluate, and deploy a wide range of open-source Large Language Models (LLMs) and Vision Language Models (VLMs). By integrating advanced features like data synthesis, hyperparameter tuning automation, and support for leading ML libraries, Oumi simplifies complex tasks, enabling efficient development and rapid deployment of custom AI models.