flashinfer-ai / flashinfer
FlashInfer: Kernel Library for LLM Serving
See what the GitHub community is most excited about this month.
FlashInfer: Kernel Library for LLM Serving
CUDA accelerated rasterization of gaussian splatting
NCCL Tests
RCCL Performance Benchmark Tests
Quantized Attention achieves speedup of 2-3x and 3-5x compared to FlashAttention and xformers, without lossing end-to-end metrics across language, image, and video models.
Causal depthwise conv1d in CUDA, with a PyTorch interface
CUDA Library Samples
cuVS - a library for vector search and clustering on the GPU
LLM training in simple, raw C/CUDA
DeepEP: an efficient expert-parallel communication library
[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl
Tile primitives for speedy kernels
cuGraph - RAPIDS Graph Analytics Library
RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.