NVIDIA / ncclLinks
Optimized primitives for collective multi-GPU communication
☆4,436Updated this week
Alternatives and similar repositories for nccl
Users that are interested in nccl are comparing it to the libraries listed below
Sorting:
- NCCL Tests☆1,423Updated this week
- Collective communications library with various primitives for multi-machine training.☆1,391Updated 3 weeks ago
- A fast GPU memory copy library based on NVIDIA GPUDirect RDMA technology☆1,338Updated last month
- oneAPI Deep Neural Network Library (oneDNN)☆3,960Updated 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…☆3,132Updated this week
- [ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl☆1,817Updated 2 years ago
- Source code examples from the Parallel Forall Blog☆1,322Updated 4 months ago
- FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/☆1,525Updated this week
- Open MPI main development repository☆2,520Updated this week
- CUDA Templates and Python DSLs for High-Performance Linear Algebra☆9,226Updated this week
- Reference implementations of MLPerf® training benchmarks☆1,739Updated last month
- Unified Communication X (mailing list - https://elist.ornl.gov/mailman/listinfo/ucx-group)☆1,569Updated this week
- Reference implementations of MLPerf® inference benchmarks☆1,525Updated this week
- CUDA Core Compute Libraries☆2,162Updated this week
- CUDA Library Samples☆2,306Updated 2 weeks ago
- Transformer related optimization, including BERT, GPT☆6,392Updated last year
- common in-memory tensor structure☆1,161Updated last week
- TorchBench is a collection of open source benchmarks used to evaluate PyTorch performance.☆1,012Updated this week
- A machine learning compiler for GPUs, CPUs, and ML accelerators☆3,956Updated this week
- ☆1,984Updated 2 years ago
- FlashInfer: Kernel Library for LLM Serving☆4,853Updated this week
- A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.☆921Updated this week
- Examples demonstrating available options to program multiple GPUs in a single node or a cluster☆864Updated 4 months ago
- A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep lear…☆5,613Updated last week
- C++ extensions in PyTorch☆1,183Updated 3 weeks ago
- The Torch-MLIR project aims to provide first class support from the PyTorch ecosystem to the MLIR ecosystem.☆1,742Updated this week
- Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training☆1,859Updated last week
- Samples for CUDA Developers which demonstrates features in CUDA Toolkit☆8,801Updated last month
- A benchmark framework for Tensorflow☆1,146Updated 2 years ago
- NVIDIA Data Center GPU Manager (DCGM) is a project for gathering telemetry and measuring the health of NVIDIA GPUs☆658Updated 2 months ago