Fault tolerance for PyTorch (HSDP, LocalSGD, DiLoCo, Streaming DiLoCo)
☆487Mar 17, 2026Updated this week
Alternatives and similar repositories for torchft
Users that are interested in torchft are comparing it to the libraries listed below
Sorting:
- NVIDIA Resiliency Extension is a python package for framework developers and users to implement fault-tolerant features. It improves the …☆271Updated this week
- A PyTorch native platform for training generative AI models☆5,162Updated this week
- 🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash…☆282Nov 24, 2025Updated 3 months ago
- PyTorch native quantization and sparsity for training and inference☆2,739Updated this week
- A performant, memory-efficient checkpointing library for PyTorch applications, designed with large, complex distributed workloads in mind…☆164Jan 12, 2026Updated 2 months ago
- Byted PyTorch Distributed for Hyperscale Training of LLMs and RLs☆1,000Mar 3, 2026Updated 2 weeks ago
- ☆21Mar 3, 2025Updated last year
- A library to analyze PyTorch traces.☆474Updated this week
- PyTorch Single Controller☆996Updated this week
- Helpful tools and examples for working with flex-attention☆1,157Feb 8, 2026Updated last month
- TorchX is a universal job launcher for PyTorch applications. TorchX is designed to have fast iteration time for training/research and sup…☆421Updated this week
- Minimalistic large language model 3D-parallelism training☆2,617Feb 19, 2026Updated last month
- Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.☆332Updated this week
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on H…☆3,231Updated this week
- Zero Bubble Pipeline Parallelism☆451May 7, 2025Updated 10 months ago
- TORCH_TRACE parser for PT2☆78Updated this week
- FlexAttention w/ FlashAttention3 Support☆27Oct 5, 2024Updated last year
- PCCL (Prime Collective Communications Library) implements fault tolerant collective communications over IP☆148Sep 12, 2025Updated 6 months ago
- Scalable and Performant Data Loading☆374Updated this week
- Fast low-bit matmul kernels in Triton☆438Feb 1, 2026Updated last month
- A subset of PyTorch's neural network modules, written in Python using OpenAI's Triton.☆598Aug 12, 2025Updated 7 months ago
- ☆124May 28, 2024Updated last year
- This repository contains the experimental PyTorch native float8 training UX☆226Aug 1, 2024Updated last year
- CUDA checkpoint and restore utility☆429Sep 15, 2025Updated 6 months ago
- A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.☆932Updated this week
- Distributed Compiler based on Triton for Parallel Systems☆1,394Mar 11, 2026Updated last week
- extensible collectives library in triton☆97Mar 31, 2025Updated 11 months ago
- Scripts for managing a large H100 cluster and fixing hardware issues to ensure smooth model training.☆323Aug 20, 2024Updated last year
- Pipeline Parallelism for PyTorch☆785Aug 21, 2024Updated last year
- Efficient Triton Kernels for LLM Training☆6,216Updated this week
- GPUd automates monitoring, diagnostics, and issue identification for GPUs☆479Updated this week
- See https://github.com/cuda-mode/triton-index/ instead!☆11May 8, 2024Updated last year
- Torch Distributed Experimental☆117Aug 5, 2024Updated last year
- TorchFix - a linter for PyTorch-using code with autofix support☆151Aug 23, 2025Updated 6 months ago
- Ring attention implementation with flash attention☆996Sep 10, 2025Updated 6 months ago
- ☆241Nov 24, 2025Updated 3 months ago
- depyf is a tool to help you understand and adapt to PyTorch compiler torch.compile.☆797Oct 13, 2025Updated 5 months ago
- torchcomms: a modern PyTorch communications API☆349Mar 16, 2026Updated last week
- ☆18Mar 18, 2024Updated 2 years ago