FlashInfer: Kernel Library for LLM Serving

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

Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, DeepSeek-R1, Qwen3, Gemma 3, TTS 2x faster with 70% less VRAM.

NMA Computational Neuroscience course

SGLang is a fast serving framework for large language models and vision language models.

Simulated experiments for "Real-Time Execution of Action Chunking Flow Policies".

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

Project Malmo is a platform for Artificial Intelligence experimentation and research built on top of Minecraft. We aim to inspire a new generation of research into challenging new problems presente…

Create Minecraft bots with a powerful, stable, and high level JavaScript API.

Cataloging released Triton kernels.

Nano vLLM

A high-throughput and memory-efficient inference and serving engine for LLMs

slime is an LLM post-training framework for RL Scaling.

A scalable asynchronous reinforcement learning implementation with in-flight weight updates.

An extremely fast Python package and project manager, written in Rust.

Ongoing research training transformer models at scale

The best ChatGPT that $100 can buy.

Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen2.5, Qwen3, Llama, and more!

Train your own SOTA deductive reasoning model

Training setup for Langchain's Open Deep Research

Our library for RL environments + evals

Async RL Training at Scale

🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org

Anemoi: A Semi-Centralized Multi-agent Systems Based on Agent-to-Agent Communication MCP server from Coral Protocol

An AI agent system for solving International Mathematical Olympiad (IMO) problems using Google's Gemini, OpenAI, and XAI APIs.

Renderer for the harmony response format to be used with gpt-oss

gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

[ICML 2025] Official PyTorch Implementation of "History-Guided Video Diffusion"