A Really Scalable RL Framework to 10k+ CPUs

PyTorch-native post-training at scale

Scaling Agentic Reinforcement Learning with a Multi-Turn, Multi-Task Framework

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

An interface library for RL post training with environments.

Build, evaluate and train General Multi-Agent Assistance with ease

(best/better) practices of megatron on veRL and tuning guide

RLinf: Reinforcement Learning Infrastructure for Embodied and Agentic AI

Cosmos-RL is a flexible and scalable Reinforcement Learning framework specialized for Physical AI applications.

A library for advanced large language model reasoning

SE-Agent is a self-evolution framework for LLM Code agents. It enables trajectory-level evolution to exchange information across reasoning paths via Revision, Recombination, and Refinement, expandi…

A library for generating difficulty-scalable, multi-tool, and verifiable agentic tasks with execution trajectories.

Verlog: A Multi-turn RL framework for LLM agents

Official repo for IRL-VLA

OpenCUA: Open Foundations for Computer-Use Agents

Feedback-Driven Tool-Use Improvements in Large Language Models via Automated Build Environments

An Open-Source Large-Scale Reinforcement Learning Project for Search Agents

Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL.

Toolchain built around the Megatron-LM for Distributed Training

MiroRL is an MCP-first reinforcement learning framework for deep research agent.

Tongyi Deep Research, the Leading Open-source Deep Research Agent

rl from zero pretrain, can it be done? yes.

SWE-Swiss: A Multi-Task Fine-Tuning and RL Recipe for High-Performance Issue Resolution

JAxtar is a project with a JAX-native implementation of parallelizeable A* & Q* solver for neural heuristic search research.