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

Open-source implementation of AlphaEvolve

Framework for evaluating and improving agents

Daytona is a Secure and Elastic Infrastructure for Running AI-Generated Code

🔥 LeetCode for PyTorch — practice implementing softmax, attention, GPT-2 and more from scratch with instant auto-grading. Jupyter-based, self-hosted or try online.

SkyRL: A Modular Full-stack RL Library for LLMs

AI-Driven Scientific and Algorithmic Discovery

Block AI-generated music on Spotify Web automatically using a community-maintained list. Runs in your browser with Tampermonkey, keeps your Spotify account free from AI artists, and updates daily.

🙌 OpenHands: AI-Driven Development

Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.

Alpha Screening with LLM Reasoning via Reinforcement Learning

🤘 TT-NN operator library, and TT-Metalium low level kernel programming model.

Information hub for our project training the largest possible historical LLMs.

The best ChatGPT that $100 can buy.

Official repository for the paper "LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code"

Code release for "Git Re-Basin: Merging Models modulo Permutation Symmetries"

verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework

[NeurIPS 2025 Spotlight] Reasoning Environments for Reinforcement Learning with Verifiable Rewards

CIFAR-10 speedruns: 94% in 2.6 seconds and 96% in 27 seconds

🌱 a fast, batteries-included static-site generator that transforms Markdown content into fully functional websites

NanoGPT (124M) in 90 seconds

The Incredible PyTorch: a curated list of tutorials, papers, projects, communities and more relating to PyTorch.

Elegant easy-to-use neural networks + scientific computing in JAX. https://docs.kidger.site/equinox/

OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games.

Projects motion of pixels to a voxel

A sub-micrometer 3D motion control plattform.

Online browser-based RTS game

Hierarchical Reasoning Model Official Release