The end of web parsing. The beginning of scalable pixel-native search. link: https://pixelrag.ai/

Sparse Lazy Array Format (SLAF) for Single-Cell Genomics

TPU inference for vLLM, with unified JAX and PyTorch support.

Training API and CLI

A collection of learning resources for curious software engineers

[NeurIPS 2025] Scaling Speculative Decoding with Lookahead Reasoning

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflo…

Implementation for FP8/INT8 Rollout for RL training without performence drop.

A simple, performant, and scalable Jax LLM!

Common recipes to run vLLM

UCCL is an efficient communication library for GPUs, covering collectives, P2P (e.g., KV cache transfer, RL weight transfer), and EP (e.g., GPU-driven)

Find secrets with Gitleaks 🔑

A collection of inspiring lists, manuals, cheatsheets, blogs, hacks, one-liners, cli/web tools and more.

aiomonitor is module that adds monitor and python REPL capabilities for asyncio application

ripgrep recursively searches directories for a regex pattern while respecting your gitignore

Deep learning for dummies. All the practical details and useful utilities that go into working with real models.

PyTorch Single Controller

Official Repo for Open-Reasoner-Zero

Build compute kernels and load them from the Hub.

A unified inference and post-training framework for accelerated video generation.

Sky-T1: Train your own O1 preview model within $450

s1: Simple test-time scaling

Scalable RL solution for advanced reasoning of language models

Hydra is a framework for elegantly configuring complex applications

DPO, but faster 🚀