PyTorch Single Controller

Contexts Optical Compression

TPU inference for vLLM, with unified JAX and PyTorch support.

An interface library for RL post training with environments.

Meta Agents Research Environments is a comprehensive platform designed to evaluate AI agents in dynamic, realistic scenarios. Unlike static benchmarks, this platform introduces evolving environment…

Post-training with Tinker

Darwin Gödel Machine: Open-Ended Evolution of Self-Improving Agents

Environments for LLM Reinforcement Learning

Easy design, testing, and deployment of optical data center networks for everyone.

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

The absolute trainer to light up AI agents.

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

The official code of ARPO & AEPO

A Production-ready Reinforcement Learning AI Agent Library brought by the Applied Reinforcement Learning team at Meta.

A programming framework for agentic AI

RAG on Everything with LEANN. Enjoy 97% storage savings while running a fast, accurate, and 100% private RAG application on your personal device.

open-source coding LLM for software engineering tasks

Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

A JAX-native LLM Post-Training Library

A fork to add multimodal model training to open-r1

Open-source implementation of AlphaEvolve

[NeurIPS '25] Challenging Software Optimization Tasks for Evaluating SWE-Agents

kernels, of the mega variety

NVIDIA Inference Xfer Library (NIXL)

Code for the paper: "Learning to Reason without External Rewards"

Scalable toolkit for efficient model reinforcement