NVlabs / instant-ngp
Instant neural graphics primitives: lightning fast NeRF and more
See what the GitHub community is most excited about today.
Instant neural graphics primitives: lightning fast NeRF and more
CUDA Library Samples
FlashInfer: Kernel Library for LLM Serving
LLM training in simple, raw C/CUDA
cuGraph - RAPIDS Graph Analytics Library
Some CUDA example code with READMEs.
[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl
[ICLR2025 Spotlight] SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models
NCCL Tests
cuVS - a library for vector search and clustering on the GPU
Quantized Attention that achieves speedups of 2.1-3.1x and 2.7-5.1x compared to FlashAttention2 and xformers, respectively, without lossing end-to-end metrics across various models.
TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing. By providing a higher-level interface, algorithm developers can design hardware-aware algorithms without dealing with low-level hardware complexities.
CUDA Kernel Benchmarking Library
Tile primitives for speedy kernels
Causal depthwise conv1d in CUDA, with a PyTorch interface
RCCL Performance Benchmark Tests