Alternatives and similar repositories for FlashAttention-PyTorch

Users that are interested in FlashAttention-PyTorch are comparing it to the libraries listed below

• mdy666 / mdy_triton
  ☆143Updated 2 months ago
• dhcode-cpp / NSA-pytorch
  DeepSeek Native Sparse Attention pytorch implementation
  ☆95Updated last month
• madsys-dev / deepseekv2-profile
  ☆147Updated 6 months ago
• fxmeng / TransMLA
  TransMLA: Multi-Head Latent Attention Is All You Need
  ☆356Updated 2 weeks ago
• FlagOpen / FlagAttention
  A collection of memory efficient attention operators implemented in the Triton language.
  ☆278Updated last year
• NVlabs / COAT
  [ICLR 2025] COAT: Compressing Optimizer States and Activation for Memory-Efficient FP8 Training
  ☆239Updated last month
• mdy666 / Qwen-Native-Sparse-Attention
  qwen-nsa
  ☆74Updated 5 months ago
• kyegomez / FlashAttention20
  Get down and dirty with FlashAttention2.0 in pytorch, plug in and play no complex CUDA kernels
  ☆109Updated 2 years ago
• qingkelab / qingketalk
  青稞Talk
  ☆145Updated last week
• xlite-dev / ffpa-attn
  🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.
  ☆217Updated last month
• OpenNLPLab / lightning-attention
  Lightning Attention-2: A Free Lunch for Handling Unlimited Sequence Lengths in Large Language Models
  ☆328Updated 6 months ago
• smart-lty / ParallelSpeculativeDecoding
  [ICLR 2025] PEARL: Parallel Speculative Decoding with Adaptive Draft Length
  ☆111Updated 5 months ago
• 66RING / tiny-flash-attention
  flash attention tutorial written in python, triton, cuda, cutlass
  ☆420Updated 4 months ago
• pprp / Awesome-Efficient-MoE
  Efficient Mixture of Experts for LLM Paper List
  ☆127Updated this week
• fanshiqing / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆146Updated 3 weeks ago
• tgale96 / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆119Updated 3 months ago
• Dao-AILab / grouped-latent-attention
  ☆126Updated 3 months ago
• harleyszhang / llm_counts
  llm theoretical performance analysis tools and support params, flops, memory and latency analysis.
  ☆106Updated 2 months ago
• kyegomez / FlashAttention20Triton
  Triton implementation of Flash Attention2.0
  ☆39Updated 2 years ago
• alexzhang13 / flashattention2-custom-mask
  Triton implementation of FlashAttention2 that adds Custom Masks.
  ☆134Updated last year
• luliyucoordinate / cute-flash-attention
  Implement Flash Attention using Cute.
  ☆95Updated 9 months ago
• feifeibear / long-context-attention
  USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference
  ☆569Updated this week
• DD-DuDa / BitDistiller
  [ACL 2024] A novel QAT with Self-Distillation framework to enhance ultra low-bit LLMs.
  ☆122Updated last year
• InternLM / Awesome-LLM-Training-System
  ☆42Updated last year
• fkodom / grouped-query-attention-pytorch
  (Unofficial) PyTorch implementation of grouped-query attention (GQA) from "GQA: Training Generalized Multi-Query Transformer Models from …
  ☆179Updated last year
• haochengxi / Train_Transformers_with_INT4
  ☆154Updated 2 years ago
• liangyuwang / Tiny-DeepSpeed
  Tiny-DeepSpeed, a minimalistic re-implementation of the DeepSpeed library
  ☆45Updated last month
• AniZpZ / AutoSmoothQuant
  An easy-to-use package for implementing SmoothQuant for LLMs
  ☆106Updated 5 months ago
• ModelTC / EasyLLM
  Built upon Megatron-Deepspeed and HuggingFace Trainer, EasyLLM has reorganized the code logic with a focus on usability. While enhancing …
  ☆49Updated last year
• ruikangliu / FlatQuant
  [ICML 2025] Official PyTorch implementation of "FlatQuant: Flatness Matters for LLM Quantization"
  ☆162Updated 2 months ago