Alternatives and similar repositories for flash-attention

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

• InternLM / turbomind
  ☆97Updated 8 months ago
• infinigence / FlashOverlap
  A lightweight design for computation-communication overlap.
  ☆188Updated last month
• AlibabaPAI / FLASHNN
  ☆102Updated last year
• infinigence / Semi-PD
  A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.
  ☆116Updated 6 months ago
• usyd-fsalab / fp6_llm
  An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).
  ☆272Updated 4 months ago
• xlite-dev / HGEMM
  ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.
  ☆134Updated 6 months ago
• microsoft / vattention
  Dynamic Memory Management for Serving LLMs without PagedAttention
  ☆444Updated 6 months ago
• LLMServe / SwiftTransformer
  High performance Transformer implementation in C++.
  ☆142Updated 10 months ago
• madsys-dev / deepseekv2-profile
  ☆152Updated 9 months ago
• feifeibear / LLMRoofline
  Compare different hardware platforms via the Roofline Model for LLM inference tasks.
  ☆119Updated last year
• stepfun-ai / StepMesh
  ☆324Updated 3 weeks ago
• Ascend / triton-ascend
  Triton adapter for Ascend. Mirror of https://gitee.com/ascend/triton-ascend
  ☆86Updated this week
• tgale96 / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆131Updated 6 months ago
• flagos-ai / FlagCX
  ☆131Updated this week
• KuangjuX / NVSHMEM-Tutorial
  NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer
  ☆144Updated 2 months ago
• DeepLink-org / DLSlime
  DLSlime: Flexible & Efficient Heterogeneous Transfer Toolkit
  ☆82Updated last week
• flashinfer-ai / cutlass-viz
  ☆65Updated 7 months ago
• alibaba / easydist
  Automated Parallelization System and Infrastructure for Multiple Ecosystems
  ☆80Updated last year
• triton-lang / kernels
  ☆94Updated last year
• xlite-dev / ffpa-attn
  🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.
  ☆233Updated 2 weeks ago
• ademeure / DeeperGEMM
  DeeperGEMM: crazy optimized version
  ☆73Updated 6 months ago
• bytedance / ByteMLPerf
  AI Accelerator Benchmark focuses on evaluating AI Accelerators from a practical production perspective, including the ease of use and ver…
  ☆276Updated 3 months ago
• AlibabaResearch / flash-llm
  Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity
  ☆224Updated 2 years ago
• tlc-pack / libflash_attn
  Standalone Flash Attention v2 kernel without libtorch dependency
  ☆112Updated last year
• HandH1998 / QQQ
  QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.
  ☆148Updated 3 months ago
• efeslab / Atom
  [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving
  ☆331Updated last year
• flagos-ai / FlagAttention
  A collection of memory efficient attention operators implemented in the Triton language.
  ☆283Updated last year
• PKU-SEC-Lab / HybriMoE
  [DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"
  ☆93Updated 5 months ago
• AlibabaPAI / torchacc
  PyTorch distributed training acceleration framework
  ☆53Updated 3 months ago
• EfficientMoE / MoE-Infinity
  PyTorch library for cost-effective, fast and easy serving of MoE models.
  ☆262Updated last month