verl-agent is an extension of veRL, designed for training LLM/VLM agents via RL. verl-agent is also the official code for paper "Group-in-Group Policy Optimization for LLM Agent Training"

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

Laser: Learn to Reason Efficiently with Adaptive Length-based Reward Shaping

Defeating the Training-Inference Mismatch via FP16

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

The Tool Decathlon: Benchmarking Language Agents for Diverse, Realistic, and Long-Horizon Task Execution

NOFX: Defining the Next-Generation AI Trading Operating System. A multi-exchange Al trading platform(Binance/Hyperliquid/Aster) with multi-Ai competition(deepseek/qwen/gemini/claude)self-evolution,…

MiniMax-M2, a model built for Max coding & agentic workflows.

BigCodeArena: Unveiling More Reliable Human Preferences in Code Generation via Execution

Post-training with Tinker

The official github repo for "Training Optimal Large Diffusion Language Models", the first-ever large-scale diffusion language models scaling law..

All-in-One Sandbox for AI Agents that combines Browser, Shell, File, MCP and VSCode Server in a single Docker container.

Checkpoint-engine is a simple middleware to update model weights in LLM inference engines

Official Code for "Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search"

End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning

[NeurIPS 2025] The official repo of SynLogic: Synthesizing Verifiable Reasoning Data at Scale for Learning Logical Reasoning and Beyond

GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning

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

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

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

The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-bench verified!

SkyRL: A Modular Full-stack RL Library for LLMs

The absolute trainer to light up AI agents.

Qwen Code is a coding agent that lives in the digital world.

Kimi K2 is the large language model series developed by Moonshot AI team