Alternatives and similar repositories for out-of-the-box-fp8-training

Users that are interested in out-of-the-box-fp8-training are comparing it to the libraries listed below

• srush / triton-autodiff
  Experiment of using Tangent to autodiff triton
  ☆81Updated last year
• drisspg / transformer_nuggets
  A place to store reusable transformer components of my own creation or found on the interwebs
  ☆60Updated 2 weeks ago
• cloneofsimo / min-fsdp
  ☆87Updated last year
• lucidrains / autoregressive-linear-attention-cuda
  CUDA implementation of autoregressive linear attention, with all the latest research findings
  ☆44Updated 2 years ago
• graphcore-research / unit-scaling
  A library for unit scaling in PyTorch
  ☆130Updated 2 months ago
• mgmalek / efficient_cross_entropy
  ☆118Updated last year
• alexzhang13 / Triton-Puzzles-Solutions
  Personal solutions to the Triton Puzzles
  ☆20Updated last year
• softmax1 / Flash-Attention-Softmax-N
  CUDA and Triton implementations of Flash Attention with SoftmaxN.
  ☆73Updated last year
• UmerHA / triton_util
  Make triton easier
  ☆47Updated last year
• HazyResearch / train-tk
  train with kittens!
  ☆62Updated 10 months ago
• amirzandieh / HyperAttention
  Triton Implementation of HyperAttention Algorithm
  ☆48Updated last year
• ezyang / torchdbg
  PyTorch centric eager mode debugger
  ☆48Updated 8 months ago
• cloneofsimo / zeroshampoo
  ☆34Updated last year
• nshepperd / flash_attn_jax
  JAX bindings for Flash Attention v2
  ☆91Updated this week
• cloneofsimo / min-max-gpt
  Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training
  ☆132Updated last year
• ClashLuke / tpucare
  Automatically take good care of your preemptible TPUs
  ☆36Updated 2 years ago
• eth-easl / fmengine
  Utilities for Training Very Large Models
  ☆58Updated 11 months ago
• meta-pytorch / float8_experimental
  This repository contains the experimental PyTorch native float8 training UX
  ☆224Updated last year
• dvruette / barrel-rec-pytorch
  ☆53Updated last year
• davisyoshida / abnormal-floats
  Code for the note "NF4 Isn't Information Theoretically Optimal (and that's Good)
  ☆20Updated 2 years ago
• lernapparat / torchhacks
  Hacks for PyTorch
  ☆19Updated 2 years ago
• lucaslingle / mu_transformer
  Transformer with Mu-Parameterization, implemented in Jax/Flax. Supports FSDP on TPU pods.
  ☆32Updated 3 months ago
• fal-ai-community / NativeSparseAttention
  research impl of Native Sparse Attention (2502.11089)
  ☆61Updated 6 months ago
• sholtodouglas / scalingExperiments
  ☆61Updated 3 years ago
• vvvm23 / mamba-jax
  Unofficial but Efficient Implementation of "Mamba: Linear-Time Sequence Modeling with Selective State Spaces" in JAX
  ☆88Updated last year
• ayaka14732 / llama-2-jax
  JAX implementation of the Llama 2 model
  ☆219Updated last year
• lucidrains / taylor-series-linear-attention
  Explorations into the recently proposed Taylor Series Linear Attention
  ☆100Updated last year
• open-lm-engine / flash-model-architectures
  A bunch of kernels that might make stuff slower 😉
  ☆58Updated 2 weeks ago
• crowsonkb / LDLM
  Latent Diffusion Language Models
  ☆69Updated last year
• proger / accelerated-scan
  Accelerated First Order Parallel Associative Scan
  ☆187Updated last year