Optimized primitives for collective multi-GPU communication
☆5,152Oct 10, 2026Updated this week
Alternatives and similar repositories for nccl
Users that are interested in nccl are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- NCCL Tests☆1,673Updated this week
- Collective communications library with various primitives for multi-machine training.☆1,458Oct 1, 2026Updated last week
- CUDA Templates and Python DSLs for High-Performance Linear Algebra☆10,557Sep 23, 2026Updated 2 weeks ago
- A fast GPU memory copy library based on NVIDIA GPUDirect RDMA technology☆1,421Sep 15, 2026Updated 3 weeks ago
- Microsoft Collective Communication Library☆400Aug 25, 2026Updated last month
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Ongoing research training transformer models at scale☆18,115Updated this week
- RDMA and SHARP plugins for nccl library☆232Aug 12, 2026Updated 2 months ago
- Development repository for the Triton language and compiler☆20,347Updated this week
- ☆402Apr 23, 2024Updated 2 years 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…☆3,573Updated this week
- Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.☆14,680Sep 17, 2026Updated 3 weeks ago
- [ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl☆1,845Oct 9, 2023Updated 3 years ago
- FlashInfer: Kernel Library for LLM Serving☆6,588Updated this week
- Open Machine Learning Compiler Framework☆13,822Updated this week
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- Unified Communication X (mailing list - https://elist.ornl.gov/mailman/listinfo/ucx-group)☆1,723Updated this week
- Distributed Compiler and Optimized Parallel Kernels☆1,562Sep 18, 2026Updated 3 weeks ago
- NVIDIA Inference Xfer Library (NIXL)☆1,296Updated this week
- NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process com…☆597Oct 3, 2026Updated last week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,370Aug 28, 2025Updated last year
- Open MPI main development repository☆2,664Updated this week
- Examples demonstrating available options to program multiple GPUs in a single node or a cluster☆916Sep 26, 2025Updated last year
- Transformer related optimization, including BERT, GPT☆6,454Mar 27, 2024Updated 2 years ago
- [DEPRECATED] Moved to ROCm/rocm-systems repo☆420Oct 2, 2026Updated last week
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Infiniband Verbs Performance Tests☆1,026Oct 1, 2026Updated last week
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆563Updated this week
- NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source compone…☆13,396Updated this week
- A high performance and generic framework for distributed DNN training☆3,717Oct 3, 2023Updated 3 years ago
- A tool for bandwidth measurements on NVIDIA GPUs.☆785Jul 28, 2026Updated 2 months ago
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆6,764Updated this week
- DeepEP: an efficient expert-parallel communication library☆10,266Sep 30, 2026Updated last week
- A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch☆9,005Updated this week
- Fast and memory-efficient exact attention☆25,115Updated this week
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- Samples for CUDA Developers which demonstrates features in CUDA Toolkit☆9,689Sep 9, 2026Updated last month
- CUDA Core Compute Libraries☆2,533Updated this week
- TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizat…☆14,793Updated this week
- This is a plugin which lets EC2 developers use libfabric as network provider while running NCCL applications.☆235Updated this week
- NCCL Fast Socket is a transport layer plugin to improve NCCL collective communication performance on Google Cloud.☆125Nov 15, 2023Updated 2 years ago
- DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.☆43,217Updated this week
- ☆608Apr 6, 2018Updated 8 years ago