Awesome Unified Multimodal Models

Fast and memory-efficient exact attention

A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.

PyTorch code and models for VJEPA2 self-supervised learning from video.

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

Fully open reproduction of DeepSeek-R1

Terraform script to deploy Outline VPN on AWS

Our library for RL environments + evals

Personal Webpage

Distributed Inference with vLLM

Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.

Build resilient language agents as graphs.

The Arcade Learning Environment (ALE) -- a platform for AI research.

Performance-Optimized AI Inference on Your GPUs. Unlock it by selecting and tuning the optimal inference engine for your model.

DeepSeek-V3, R1 671B on 8xH100 Throughput Benchmarks

A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.

Official code repo for the O'Reilly Book - "Hands-On Large Language Models"

PyTorch emulation library for Microscaling (MX)-compatible data formats

SGLang is a high-performance serving framework for large language models and multimodal models.

Official release of InternLM series (InternLM, InternLM2, InternLM2.5, InternLM3).

Code for BLT research paper

Interactive visualizations of the geometric intuition behind diffusion models.

Presentation Slides for Developers

An explainable and simplified version of OLMo model

verl: Volcano Engine Reinforcement Learning for LLMs

Tools to Design or Visualize Architecture of Neural Network

Minimal and annotated implementations of key ideas from modern deep learning research.

🚀 Efficient implementations of state-of-the-art linear attention models