Stockfish NNUE (Chess evaluation) trainer in Pytorch

Toward High-Accuracy Open-Source Biomolecular Structure Prediction.

A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.

Specialised ML Library

A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.

pgf/TikZ diagram editor

llama.cpp fork with additional SOTA quants and improved performance

Provide with pre-build flash-attention 2 and 3 package wheels on Linux and Windows using GitHub Actions

Code for "ODESteer: A Unified ODE-Based Steering Framework for LLM Alignment"

ENACT is a benchmark that evaluates embodied cognition through world modeling from egocentric interaction. It is designed to be simple and have a scalable dataset.

This is the code for AttentionPack (CVPR 2026)

OpenShell is the safe, private runtime for autonomous AI agents.

An open-source AI agent that lives in your terminal.

CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation

TinyChatEngine: On-Device LLM Inference Library

Everyone's favourite kernel generation agent, now in your CLI!

A declarative inline TUI rendering library for Rust, built on Ratatui

✨ Making your shell magical

A fast, cross-platform build tool inspired by Make, designed for modern workflows.

🦉 Data Versioning and ML Experiments

The open source AI engineering platform for agents, LLMs, and ML models. MLflow enables teams of all sizes to debug, evaluate, monitor, and optimize production-quality AI applications while control…

Hydra is a framework for elegantly configuring complex applications

BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.

Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.

Open source web interface for Slurm HPC & AI clusters

OpenViking is an open-source context database designed specifically for AI Agents(such as openclaw). OpenViking unifies the management of context (memory, resources, and skills) that Agents need th…

PyTorch native quantization and sparsity for training and inference

An acceleration library that supports arbitrary bit-width combinatorial quantization operations

IntelliJ-like CMD+Shift+F in Cursor / VSCode.