Alternatives and similar repositories for flash-attention

Users that are interested in flash-attention are comparing it to the libraries listed below

• ROCm / triton
  Development repository for the Triton language and compiler
  ☆126Updated this week
• usyd-fsalab / fp6_llm
  An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).
  ☆260Updated 2 weeks ago
• microsoft / BitBLAS
  BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.
  ☆654Updated 3 weeks ago
• fpgaminer / GPTQ-triton
  GPTQ inference Triton kernel
  ☆303Updated 2 years ago
• mobiusml / gemlite
  Fast low-bit matmul kernels in Triton
  ☆338Updated last week
• SqueezeAILab / KVQuant
  [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization
  ☆365Updated 11 months ago
• efeslab / Atom
  [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving
  ☆318Updated last year
• ROCm / aiter
  AI Tensor Engine for ROCm
  ☆243Updated this week
• neuralmagic / compressed-tensors
  A safetensors extension to efficiently store sparse quantized tensors on disk
  ☆141Updated this week
• ROCm / vllm
  A high-throughput and memory-efficient inference and serving engine for LLMs
  ☆86Updated this week
• pytorch-labs / applied-ai
  Applied AI experiments and examples for PyTorch
  ☆289Updated 2 months ago
• efeslab / fiddler
  [ICLR'25] Fast Inference of MoE Models with CPU-GPU Orchestration
  ☆224Updated 8 months ago
• wangsiping97 / FastGEMV
  High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.
  ☆113Updated last year
• mit-han-lab / omniserve
  [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…
  ☆730Updated 4 months ago
• nbasyl / LLM-FP4
  The official implementation of the EMNLP 2023 paper LLM-FP4
  ☆211Updated last year
• neuralmagic / AutoFP8
  ☆195Updated 2 months ago
• HandH1998 / QQQ
  QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.
  ☆137Updated 3 months ago
• FlagOpen / FlagAttention
  A collection of memory efficient attention operators implemented in the Triton language.
  ☆273Updated last year
• Dao-AILab / quack
  A Quirky Assortment of CuTe Kernels
  ☆374Updated this week
• IST-DASLab / marlin
  FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.
  ☆868Updated 11 months ago
• Dao-AILab / fast-hadamard-transform
  Fast Hadamard transform in CUDA, with a PyTorch interface
  ☆213Updated last year
• ROCm / aotriton
  Ahead of Time (AOT) Triton Math Library
  ☆74Updated last week
• foundation-model-stack / foundation-model-stack
  🚀 Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.
  ☆206Updated last week
• triton-lang / kernels
  ☆85Updated 8 months ago
• ColfaxResearch / cutlass-kernels
  ☆227Updated last year
• Cornell-RelaxML / qtip
  ☆143Updated last month
• spcl / QuaRot
  Code for Neurips24 paper: QuaRot, an end-to-end 4-bit inference of large language models.
  ☆410Updated 8 months ago
• IST-DASLab / Sparse-Marlin
  Boosting 4-bit inference kernels with 2:4 Sparsity
  ☆80Updated 11 months ago
• hahnyuan / RPTQ4LLM
  Reorder-based post-training quantization for large language model
  ☆193Updated 2 years ago
• stanford-futuredata / stk
  ☆107Updated 11 months ago