🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash attention v2.
☆284Nov 24, 2025Updated 4 months ago
Alternatives and similar repositories for fms-fsdp
Users that are interested in fms-fsdp are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- 🚀 Collection of components for development, training, tuning, and inference of foundation models leveraging PyTorch native components.☆222Apr 3, 2026Updated last week
- ☆25Sep 9, 2024Updated last year
- Applied AI experiments and examples for PyTorch☆320Aug 22, 2025Updated 7 months ago
- 🚀 Collection of libraries used with fms-hf-tuning to accelerate fine-tuning and training of large models.☆14Jan 30, 2026Updated 2 months ago
- A PyTorch native platform for training generative AI models☆5,205Apr 3, 2026Updated last week
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- Fault tolerance for PyTorch (HSDP, LocalSGD, DiLoCo, Streaming DiLoCo)☆491Apr 3, 2026Updated last week
- Minimalistic large language model 3D-parallelism training☆2,632Apr 2, 2026Updated last week
- ☆93Jul 5, 2024Updated last year
- Pipeline Parallelism for PyTorch☆785Aug 21, 2024Updated last year
- Minimalistic 4D-parallelism distributed training framework for education purpose☆2,137Aug 26, 2025Updated 7 months ago
- Ring attention implementation with flash attention☆1,003Sep 10, 2025Updated 7 months ago
- Parallel Associative Scan for Language Models☆18Jan 8, 2024Updated 2 years ago
- Legible, Scalable, Reproducible Foundation Models with Named Tensors and Jax☆703Jan 26, 2026Updated 2 months ago
- ☆20May 30, 2024Updated last year
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- ☆19Dec 4, 2025Updated 4 months ago
- PyTorch native quantization and sparsity for training and inference☆2,756Updated this week
- ☆34Sep 10, 2024Updated last year
- 🚀 Efficient implementations for emerging model architectures☆4,823Updated this week
- Checkpointable dataset utilities for foundation model training☆32Jan 29, 2024Updated 2 years ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆660Jan 15, 2026Updated 2 months ago
- Helpful tools and examples for working with flex-attention☆1,167Apr 1, 2026Updated last week
- ☆124May 28, 2024Updated last year
- Byted PyTorch Distributed for Hyperscale Training of LLMs and RLs☆1,004Mar 3, 2026Updated last month
- Wordpress hosting with auto-scaling on Cloudways • AdFully Managed hosting built for WordPress-powered businesses that need reliable, auto-scalable hosting. Cloudways SafeUpdates now available.
- Multipack distributed sampler for fast padding-free training of LLMs☆207Aug 10, 2024Updated last year
- Large Context Attention☆770Oct 13, 2025Updated 5 months ago
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆132Apr 17, 2024Updated last year
- Triton-based implementation of Sparse Mixture of Experts.☆272Oct 3, 2025Updated 6 months ago
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on H…☆3,256Updated this week
- FlashInfer: Kernel Library for LLM Serving☆5,273Updated this week
- 🚀 Collection of tuning recipes with HuggingFace SFTTrainer and PyTorch FSDP.☆56Mar 30, 2026Updated last week
- Efficient Triton Kernels for LLM Training☆6,265Updated this week
- A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.☆599Aug 12, 2025Updated 7 months ago
- DigitalOcean Gradient AI Platform • AdBuild production-ready AI agents using customizable tools or access multiple LLMs through a single endpoint. Create custom knowledge bases or connect external data.
- Zero Bubble Pipeline Parallelism☆452May 7, 2025Updated 11 months ago
- Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.☆4,757Jul 18, 2025Updated 8 months ago
- Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.☆6,190Aug 22, 2025Updated 7 months ago
- ☆20Jul 12, 2023Updated 2 years ago
- FlexAttention w/ FlashAttention3 Support☆27Oct 5, 2024Updated last year
- Efficient PScan implementation in PyTorch☆17Jan 2, 2024Updated 2 years ago
- PyTorch native post-training library☆5,720Apr 3, 2026Updated last week