Materials for learning SGLang

NVIDIA curated collection of educational resources related to general purpose GPU programming.

Sumi | The latest, best and especially most performant Nintendo Switch emulator! Run Nintendo Switch titles on your Android, Windows, Mac and Linux devices :)

A highly-flexible GPU simulator for AMD GPUs.

ASTRA-sim2.0: Modeling Hierarchical Networks and Disaggregated Systems for Large-model Training at Scale

A curated collection of resources, tutorials, and best practices for learning and mastering NVIDIA CUTLASS

LLM Inference analyzer for different hardware platforms

Repository to host and maintain scale-sim-v2 code

A tool for bandwidth measurements on NVIDIA GPUs.

Fastest kernels written from scratch

Timeloop performs modeling, mapping and code-generation for tensor algebra workloads on various accelerator architectures.

DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling

CUDA Templates for Linear Algebra Subroutines

Unofficial description of the CUDA assembly (SASS) instruction sets.

FlashInfer: Kernel Library for LLM Serving

DeepEP: an efficient expert-parallel communication library

SGLang is a fast serving framework for large language models and vision language models.

Material for gpu-mode lectures

Source code that accompanies The CUDA Handbook.

Performance Prediction Toolkit

📚Tensor/CUDA Cores, 📖150+ CUDA Kernels, ⚡️⚡️toy-hgemm library with WMMA, MMA and CuTe (98%~100% TFLOPS of cuBLAS 🎉🎉).