Alternatives and similar repositories for nccl

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

• lianakoleva / no-libtorch-compile
  ☆21Updated 9 months ago
• imbue-ai / cluster-health
  ☆317Updated last year
• thecharlieblake / lovely-llama
  An implementation of the Llama architecture, to instruct and delight
  ☆21Updated 6 months ago
• meta-pytorch / autoparallel
  An experimental implementation of compiler-driven automatic sharding of models across a given device mesh.
  ☆47Updated this week
• HazyResearch / train-tk
  train with kittens!
  ☆63Updated last year
• ezyang / torchdbg
  PyTorch centric eager mode debugger
  ☆48Updated 11 months ago
• NVIDIA / nvidia-resiliency-ext
  NVIDIA Resiliency Extension is a python package for framework developers and users to implement fault-tolerant features. It improves the …
  ☆239Updated this week
• cat-state / tinypar
  ☆20Updated 2 years ago
• meta-pytorch / torchsnapshot
  A performant, memory-efficient checkpointing library for PyTorch applications, designed with large, complex distributed workloads in mind…
  ☆161Updated 2 months ago
• coreweave / ml-containers
  ☆42Updated this week
• lucaslingle / mu_transformer
  Transformer with Mu-Parameterization, implemented in Jax/Flax. Supports FSDP on TPU pods.
  ☆32Updated 6 months ago
• open-lm-engine / accelerated-model-architectures
  A bunch of kernels that might make stuff slower 😉
  ☆65Updated last week
• graphcore-research / jax-scalify
  JAX Scalify: end-to-end scaled arithmetics
  ☆17Updated last year
• mgmalek / efficient_cross_entropy
  ☆121Updated last year
• vdesai2014 / inference-optimization-blog-post
  ☆90Updated last year
• fattorib / ZeRO-transformer
  Two implementations of ZeRO-1 optimizer sharding in JAX
  ☆14Updated 2 years ago
• meta-pytorch / torchcomms
  torchcomms: a modern PyTorch communications API
  ☆302Updated this week
• meta-pytorch / float8_experimental
  This repository contains the experimental PyTorch native float8 training UX
  ☆227Updated last year
• gpu-mode / ring-attention
  ring-attention experiments
  ☆160Updated last year
• open-lm-engine / lm-engine
  LM engine is a library for pretraining/finetuning LLMs
  ☆77Updated last week
• Jaykef / Triton-nanoGPT
  Custom triton kernels for training Karpathy's nanoGPT.
  ☆19Updated last year
• srush / triton-autodiff
  Experiment of using Tangent to autodiff triton
  ☆81Updated last year
• meta-pytorch / torchft
  Fault tolerance for PyTorch (HSDP, LocalSGD, DiLoCo, Streaming DiLoCo)
  ☆456Updated last week
• zinccat / Awesome-Triton-Kernels
  Collection of kernels written in Triton language
  ☆173Updated 8 months ago
• gpu-mode / discord-cluster-manager
  Write a fast kernel and run it on Discord. See how you compare against the best!
  ☆64Updated 2 weeks ago
• graphcore-research / unit-scaling
  A library for unit scaling in PyTorch
  ☆132Updated 5 months ago
• graphcore-research / out-of-the-box-fp8-training
  Demo of the unit_scaling library, showing how a model can be easily adapted to train in FP8.
  ☆46Updated last year
• foundation-model-stack / fms-fsdp
  🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash…
  ☆271Updated 3 weeks ago
• cloneofsimo / min-fsdp
  ☆91Updated last year
• siboehm / ShallowSpeed
  Small scale distributed training of sequential deep learning models, built on Numpy and MPI.
  ☆153Updated 2 years ago