Alternatives and similar repositories for flash-attention

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

• InternLM / turbomind
  ☆87Updated 3 months ago
• infinigence / FlashOverlap
  A lightweight design for computation-communication overlap.
  ☆146Updated 3 weeks ago
• feifeibear / LLMRoofline
  Compare different hardware platforms via the Roofline Model for LLM inference tasks.
  ☆107Updated last year
• usyd-fsalab / fp6_llm
  An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).
  ☆254Updated 8 months ago
• infinigence / Semi-PD
  A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.
  ☆93Updated last month
• AlibabaPAI / FLASHNN
  ☆96Updated 10 months ago
• madsys-dev / deepseekv2-profile
  ☆142Updated 4 months ago
• bytedance / ByteMLPerf
  AI Accelerator Benchmark focuses on evaluating AI Accelerators from a practical production perspective, including the ease of use and ver…
  ☆249Updated 2 weeks ago
• microsoft / vattention
  Dynamic Memory Management for Serving LLMs without PagedAttention
  ☆401Updated last month
• microsoft / sarathi-serve
  A low-latency & high-throughput serving engine for LLMs
  ☆388Updated last month
• xlite-dev / HGEMM
  ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.
  ☆86Updated 2 months ago
• FlagOpen / FlagAttention
  A collection of memory efficient attention operators implemented in the Triton language.
  ☆272Updated last year
• flashinfer-ai / cutlass-viz
  ☆60Updated 2 months ago
• AlibabaResearch / flash-llm
  Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity
  ☆216Updated last year
• tgale96 / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆100Updated last month
• LLMServe / SwiftTransformer
  High performance Transformer implementation in C++.
  ☆125Updated 5 months ago
• triton-lang / kernels
  ☆83Updated 8 months ago
• xlite-dev / ffpa-attn
  ⚡️FFPA: Extend FlashAttention-2 with Split-D, achieve ~O(1) SRAM complexity for large headdim, 1.8x~3x↑ vs SDPA.🎉
  ☆189Updated 2 months ago
• ademeure / DeeperGEMM
  DeeperGEMM: crazy optimized version
  ☆69Updated 2 months ago
• alibaba / easydist
  Automated Parallelization System and Infrastructure for Multiple Ecosystems
  ☆79Updated 7 months ago
• Azure / msccl-executor-nccl
  ☆37Updated 7 months ago
• HandH1998 / QQQ
  QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.
  ☆133Updated 3 months ago
• tlc-pack / libflash_attn
  Standalone Flash Attention v2 kernel without libtorch dependency
  ☆110Updated 10 months ago
• yifuwang / symm-mem-recipes
  ☆94Updated 6 months ago
• Bruce-Lee-LY / flash_attention_inference
  Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.
  ☆39Updated 4 months ago
• EfficientMoE / MoE-Infinity
  PyTorch library for cost-effective, fast and easy serving of MoE models.
  ☆207Updated this week
• efeslab / Atom
  [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving
  ☆313Updated last year
• tlc-pack / cutlass_fpA_intB_gemm
  A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer
  ☆92Updated last week
• pytorch-labs / tritonbench
  Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.
  ☆176Updated last week
• neuralmagic / AutoFP8
  ☆195Updated 2 months ago