Alternatives and similar repositories for torch_utils

Users that are interested in torch_utils are comparing it to the libraries listed below

• AlibabaPAI / FLASHNN
  ☆102Updated last year
• tgale96 / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆131Updated 6 months ago
• fanshiqing / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆169Updated last month
• fzyzcjy / torch_memory_saver
  Allow torch tensor memory to be released and resumed later
  ☆177Updated this week
• infinigence / FlashOverlap
  A lightweight design for computation-communication overlap.
  ☆190Updated last month
• luliyucoordinate / cute-flash-attention
  Implement Flash Attention using Cute.
  ☆97Updated 11 months ago
• flashinfer-ai / cutlass-viz
  ☆65Updated 7 months ago
• InternLM / Awesome-LLM-Training-System
  ☆44Updated last year
• microsoft / nnscaler
  nnScaler: Compiling DNN models for Parallel Training
  ☆120Updated 2 months ago
• flagos-ai / FlagAttention
  A collection of memory efficient attention operators implemented in the Triton language.
  ☆285Updated last year
• yifuwang / symm-mem-recipes
  ☆148Updated 11 months ago
• thunlp / Seq1F1B
  Sequence-level 1F1B schedule for LLMs.
  ☆37Updated 3 months ago
• kwai / Megatron-Kwai
  LLM training technologies developed by kwai
  ☆66Updated last week
• KuangjuX / NVSHMEM-Tutorial
  NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer
  ☆144Updated 2 months ago
• stepfun-ai / StepMesh
  ☆324Updated 3 weeks ago
• InternLM / turbomind
  ☆97Updated 8 months ago
• nex-agi / NexVenusCL
  Nex Venus Communication Library
  ☆59Updated 2 weeks ago
• madsys-dev / deepseekv2-profile
  ☆152Updated 9 months ago
• Victarry / PP-Schedule-Visualization
  Pipeline Parallelism Emulation and Visualization
  ☆72Updated 5 months ago
• CalvinXKY / mfu_calculation
  A simple calculation for LLM MFU.
  ☆50Updated 2 months ago
• ademeure / DeeperGEMM
  DeeperGEMM: crazy optimized version
  ☆73Updated 6 months ago
• CalebDu / Awesome-Cute
  ☆112Updated 6 months ago
• osayamenja / FlashMoE
  Distributed MoE in a Single Kernel [NeurIPS '25]
  ☆149Updated last week
• microsoft / chunk-attention
  ☆82Updated 7 months ago
• AlibabaResearch / flash-llm
  Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity
  ☆224Updated 2 years ago
• alibaba / easydist
  Automated Parallelization System and Infrastructure for Multiple Ecosystems
  ☆80Updated last year
• RulinShao / LightSeq
  Official repository for DistFlashAttn: Distributed Memory-efficient Attention for Long-context LLMs Training
  ☆218Updated last year
• meta-pytorch / KernelAgent
  Autonomous GPU Kernel Generation via Deep Agents
  ☆163Updated last week
• yanring / Megatron-MoE-ModelZoo
  Best practices for training DeepSeek, Mixtral, Qwen and other MoE models using Megatron Core.
  ☆130Updated 3 weeks ago
• zhuohan123 / terapipe
  ☆77Updated 4 years ago