Planner/Generator/Evaluator orchestration harness for Claude Code (and Codex)

Ideas → Roadmap → Steps — a planning plugin for Claude Code

FlashInfer: Kernel Library for LLM Serving

NCCL communication API layer, and transport layer created from first principles.

NCCL Tests

A Quirky Assortment of CuTe Kernels

Optimized primitives for collective multi-GPU communication

cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.

A nvImageCodec library of GPU- and CPU- accelerated codecs featuring a unified interface

TRaSH-Guides is a comprehensive collection of guides for Radarr, Sonarr, and related media management applications.

A list of awesome compiler projects and papers for tensor computation and deep learning.

The official PyTorch implementation of the paper "Human Motion Diffusion Model"

CUDA Templates and Python DSLs for High-Performance Linear Algebra

RTX compute samples

[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl