tspeterkim / flash-attention-minimalLinks
Flash Attention in ~100 lines of CUDA (forward pass only)
☆1,044Updated last year
Alternatives and similar repositories for flash-attention-minimal
Users that are interested in flash-attention-minimal are comparing it to the libraries listed below
Sorting:
- Fast CUDA matrix multiplication from scratch☆1,011Updated 4 months ago
- A Quirky Assortment of CuTe Kernels☆749Updated this week
- Mirage Persistent Kernel: Compiling LLMs into a MegaKernel☆2,065Updated last week
- kernels, of the mega variety☆648Updated 3 months ago
- A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.☆594Updated 5 months ago
- Tile primitives for speedy kernels☆3,075Updated this week
- flash attention tutorial written in python, triton, cuda, cutlass☆475Updated 8 months ago
- Puzzles for learning Triton☆2,226Updated last year
- Distributed Compiler based on Triton for Parallel Systems☆1,313Updated 3 weeks ago
- Yet Another Language Model: LLM inference in C++/CUDA, no libraries except for I/O☆547Updated 4 months ago
- Fastest kernels written from scratch☆523Updated 4 months ago
- Step-by-step optimization of CUDA SGEMM☆422Updated 3 years ago
- BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.☆743Updated 5 months ago
- Puzzles for learning Triton, play it with minimal environment configuration!☆590Updated 3 weeks ago
- ☆271Updated last week
- GPU programming related news and material links☆1,906Updated 4 months ago
- A throughput-oriented high-performance serving framework for LLMs☆937Updated 2 months ago
- Cataloging released Triton kernels.☆287Updated 4 months ago
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆802Updated 10 months ago
- depyf is a tool to help you understand and adapt to PyTorch compiler torch.compile.☆778Updated 3 months ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆982Updated last year
- KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)☆763Updated this week
- Ring attention implementation with flash attention☆963Updated 4 months ago
- FlagGems is an operator library for large language models implemented in the Triton Language.☆871Updated this week
- Fast low-bit matmul kernels in Triton☆423Updated last month
- Several optimization methods of half-precision general matrix multiplication (HGEMM) using tensor core with WMMA API and MMA PTX instruct…☆515Updated last year
- Perplexity GPU Kernels☆552Updated 2 months ago
- A Easy-to-understand TensorOp Matmul Tutorial☆404Updated last week
- A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.☆711Updated this week
- LLM KV cache compression made easy☆749Updated last month