Stars
A flagship 560-billion-parameter open-source MoE model that advances Native Formal Reasoning in Lean4 through agentic tool-integrated reasoning.
Lightning-Fast RL for LLM Reasoning and Agents. Made Simple & Flexible.
OpenClaw-RL: Train any agent simply by talking
"🐈 nanobot: The Ultra-Lightweight OpenClaw"
Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
MiniMax-M2, a model built for Max coding & agentic workflows.
Fast, Sharp & Reliable Agentic Intelligence
gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI
MiniMax M2.1, a SOTA model for real-world dev & agents.
This is the official repo for the paper "LongCat-Flash-Omni Technical Report"
[ICLR'26] R-HORIZON: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?
LongCat Audio Tokenizer and Detokenizer
The offical repo for "Parallel-R1: Towards Parallel Thinking via Reinforcement Learning"
[ICLR'2026] R-HORIZON: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?
MiniMax-M1, the world's first open-weight, large-scale hybrid-attention reasoning model.
Latest Advances on System-2 Reasoning
Unleashing the Power of Reinforcement Learning for Math and Code Reasoners
[ICLR 2026] Code, benchmark and environment for "ScienceBoard: Evaluating Multimodal Autonomous Agents in Realistic Scientific Workflows"
[EMNLP2025 Main] Code, Result and Files for paper[Do Large Language Models Excel in Complex Logical Reasoning with Formal Language?]
[NeurIPS 2025] Reinforcement Learning for Reasoning in Large Language Models with One Training Example
Technical report of Kimina-Prover Preview.