LLM training in simple, raw C/CUDA

Instant neural graphics primitives: lightning fast NeRF and more

The project is an official implementation of our CVPR2019 paper "Deep High-Resolution Representation Learning for Human Pose Estimation"

Fast parallel CTC.

Squeeze-and-Excitation Networks

[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.

Tile primitives for speedy kernels

Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

GPU Accelerated t-SNE for CUDA with Python bindings

Fully Convolutional Instance-aware Semantic Segmentation

Deformable ConvNets V2 (DCNv2) in PyTorch

Efficient GPU kernels for block-sparse matrix multiplication and convolution

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 …

[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.

Code from the "CUDA Crash Course" YouTube series by CoffeeBeforeArch

Pytorch Bindings for warp-ctc

Reference implementation of real-time autoregressive wavenet inference

Distribution-Aware Coordinate Representation for Human Pose Estimation

A GPU implementation of Convolutional Neural Nets in C++

PopSift is an implementation of the SIFT algorithm in CUDA.

Integral Human Pose Regression

A UNIVERSAL MUSIC TRANSLATION NETWORK - a method for translating music across musical instruments and styles.

PyTorch implementation of Deformable Convolution

Official pytorch Code for CVPR2019 paper "Fast Human Pose Estimation" https://arxiv.org/abs/1811.05419

GPU implementation of a fast generalized ANS (asymmetric numeral system) entropy encoder and decoder, with extensions for lossless compression of numerical and other data types in HPC/ML applications.

Chamfer Distance in Pytorch with f-score

Approximate nearest neighbor search with product quantization on GPU in pytorch and cuda

Implementation of fused cosine similarity attention in the same style as Flash Attention

CUDA Matrix Factorization Library with Alternating Least Square (ALS)

GGNN: State of the Art Graph-based GPU Nearest Neighbor Search