GPU
[ARCHIVED] The C++ parallel algorithms library. See https://github.com/NVIDIA/cccl
Comparsion of Julia's GPU Kernel based ODE solvers with other open-source GPU ODE solvers
Collection of common code that's shared among different research projects in FAIR computer vision team.
Interpolation and function approximation with JAX
A collection of memory efficient attention operators implemented in the Triton language.
Triton Implementation of HyperAttention Algorithm
how to optimize some algorithm in cuda.
GPU-based first-order solver for linear programming.
A simple and efficient Mamba implementation in pure PyTorch and MLX.
Flash Attention in ~100 lines of CUDA (forward pass only)
Finetune Qwen3, Llama 4, TTS, DeepSeek-R1 & Gemma 3 LLMs 2x faster with 70% less memory! 🦥
Helpful tools and examples for working with flex-attention
We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel template library designed to elevate CUDA C’s level of abstra…
Mirage: Automatically Generating Fast GPU Kernels without Programming in Triton/CUDA
Tile primitives for speedy kernels
Efficient Triton Kernels for LLM Training
cuEquivariance is a math library that is a collective of low-level primitives and tensor ops to accelerate widely-used models, like DiffDock, MACE, Allegro and NEQUIP, based on equivariant neural n…
📚LeetCUDA: Modern CUDA Learn Notes with PyTorch for Beginners🐑, 200+ CUDA/Tensor Cores Kernels, HGEMM, FA-2 MMA etc.🔥
KernelBench: Can LLMs Write GPU Kernels? - Benchmark with Torch -> CUDA problems
A curated collection of resources, tutorials, and best practices for learning and mastering NVIDIA CUTLASS
FlagGems is an operator library for large language models implemented in the Triton Language.
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
Distributed Triton for Parallel Systems