Alternatives and similar repositories for SOAP

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

• cloneofsimo / zeroshampoo
  ☆34Updated last year
• amirzandieh / HyperAttention
  Triton Implementation of HyperAttention Algorithm
  ☆48Updated last year
• warner-benjamin / optimi
  Fast, Modern, and Low Precision PyTorch Optimizers
  ☆109Updated last week
• ethansmith2000 / fsdp_optimizers
  supporting pytorch FSDP for optimizers
  ☆84Updated 9 months ago
• epfml / DenseFormer
  ☆82Updated last year
• main-horse / hnet
  H-Net Dynamic Hierarchical Architecture
  ☆79Updated this week
• ethansmith2000 / TransformerExperiments
  ☆19Updated 4 months ago
• dvruette / barrel-rec-pytorch
  ☆53Updated last year
• fal-ai-community / NativeSparseAttention
  research impl of Native Sparse Attention (2502.11089)
  ☆61Updated 6 months ago
• BlinkDL / SmallInitEmb
  LayerNorm(SmallInit(Embedding)) in a Transformer to improve convergence
  ☆59Updated 3 years ago
• lucidrains / grokfast-pytorch
  Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"
  ☆101Updated 8 months ago
• euclaise / supertrainer2000
  ☆49Updated last year
• cloneofsimo / min-max-gpt
  Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training
  ☆132Updated last year
• evanatyourservice / llm-jax
  Train a SmolLM-style llm on fineweb-edu in JAX/Flax with an assortment of optimizers.
  ☆19Updated last month
• ClashLuke / tpucare
  Automatically take good care of your preemptible TPUs
  ☆36Updated 2 years ago
• shikaiqiu / compute-better-spent
  ☆57Updated 11 months ago
• cloneofsimo / min-fsdp
  ☆88Updated last year
• joey00072 / ohara
  Collection of autoregressive model implementation
  ☆86Updated 4 months ago
• recursal / GoldFinch-paper
  GoldFinch and other hybrid transformer components
  ☆45Updated last year
• LucasPrietoAl / grokking-at-the-edge-of-numerical-stability
  ☆102Updated last month
• bloc97 / DeMo
  DeMo: Decoupled Momentum Optimization
  ☆190Updated 9 months ago
• lucidrains / quartic-transformer
  Exploring an idea where one forgets about efficiency and carries out attention across each edge of the nodes (tokens)
  ☆53Updated 5 months ago
• EleutherAI / improved-t5
  Experiments for efforts to train a new and improved t5
  ☆76Updated last year
• lucidrains / gateloop-transformer
  Implementation of GateLoop Transformer in Pytorch and Jax
  ☆90Updated last year
• crowsonkb / LDLM
  Latent Diffusion Language Models
  ☆69Updated last year
• lucidrains / pause-transformer
  Yet another random morning idea to be quickly tried and architecture shared if it works; to allow the transformer to pause for any amount…
  ☆53Updated last year
• fal-ai-community / nano-mdm
  Tiny re-implementation of MDM in style of LLaDA and nano-gpt speedrun
  ☆56Updated 6 months ago
• graphcore-research / jax-scalify
  JAX Scalify: end-to-end scaled arithmetics
  ☆16Updated 10 months ago
• epfml / schedules-and-scaling
  Code for NeurIPS 2024 Spotlight: "Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations"
  ☆82Updated 10 months ago
• huyphan168 / PEER
  Mixture of A Million Experts
  ☆47Updated last year