flashinfer-ai / flashinfer
FlashInfer: Kernel Library for LLM Serving
See what the GitHub community is most excited about today.
FlashInfer: Kernel Library for LLM Serving
CUDA Library Samples
[ICLR2025 Spotlight] SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models
NCCL Tests
CUDA Kernel Benchmarking Library
[ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl
Instant neural graphics primitives: lightning fast NeRF and more
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.
cuVS - a library for vector search and clustering on the GPU
Causal depthwise conv1d in CUDA, with a PyTorch interface
Tile primitives for speedy kernels
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.
FSA/FST algorithms, differentiable, with PyTorch compatibility.