Alternatives and similar repositories for Megatron-LM

Users that are interested in Megatron-LM are comparing it to the libraries listed below

• kwai / Megatron-Kwai
  [USENIX ATC '24] Accelerating the Training of Large Language Models using Efficient Activation Rematerialization and Optimal Hybrid Paral…
  ☆65Updated last year
• AlibabaPAI / FLASHNN
  ☆100Updated last year
• Victarry / PP-Schedule-Visualization
  Pipeline Parallelism Emulation and Visualization
  ☆68Updated 4 months ago
• infinigence / FlashOverlap
  A lightweight design for computation-communication overlap.
  ☆181Updated last week
• microsoft / nnscaler
  nnScaler: Compiling DNN models for Parallel Training
  ☆118Updated last month
• CalebDu / Awesome-Cute
  ☆107Updated 5 months ago
• thunlp / Seq1F1B
  Sequence-level 1F1B schedule for LLMs.
  ☆32Updated last month
• fanshiqing / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆154Updated last week
• madsys-dev / deepseekv2-profile
  ☆148Updated 7 months ago
• weishengying / cutlass_flash_atten_fp8
  使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention
  ☆76Updated last year
• yifuwang / symm-mem-recipes
  ☆141Updated 9 months ago
• KuangjuX / NVSHMEM-Tutorial
  NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer
  ☆138Updated last month
• infinigence / Semi-PD
  A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.
  ☆112Updated 5 months ago
• OpenPPL / ppl.llm.kernel.cuda
  ☆150Updated 9 months ago
• ConnollyLeon / awesome-Auto-Parallelism
  A baseline repository of Auto-Parallelism in Training Neural Networks
  ☆147Updated 3 years ago
• reed-lau / cute-gemm
  ☆137Updated 10 months ago
• LLMServe / SwiftTransformer
  High performance Transformer implementation in C++.
  ☆138Updated 9 months ago
• tlc-pack / cutlass_fpA_intB_gemm
  A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer
  ☆94Updated last month
• tgale96 / grouped_gemm
  PyTorch bindings for CUTLASS grouped GEMM.
  ☆124Updated 4 months ago
• ParCIS / Chimera
  Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.
  ☆67Updated 7 months ago
• alibaba / easydist
  Automated Parallelization System and Infrastructure for Multiple Ecosystems
  ☆80Updated 11 months ago
• pzhao-eng / FlashMLA
  ☆56Updated 3 months ago
• DD-DuDa / Cute-Learning
  Examples of CUDA implementations by Cutlass CuTe
  ☆241Updated 3 months ago
• FlagOpen / FlagAttention
  A collection of memory efficient attention operators implemented in the Triton language.
  ☆282Updated last year
• fzyzcjy / torch_memory_saver
  Allow torch tensor memory to be released and resumed later
  ☆150Updated this week
• OpenPPL / ppl.nn.llm
  ☆139Updated last year
• Karbo123 / pytorch_grouped_gemm
  High Performance Grouped GEMM in PyTorch
  ☆31Updated 3 years ago
• AlibabaResearch / flash-llm
  Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity
  ☆221Updated 2 years ago
• stepfun-ai / StepMesh
  ☆307Updated 3 weeks ago
• AlibabaPAI / torchacc
  PyTorch distributed training acceleration framework
  ☆53Updated 2 months ago