NVIDIA / ncclLinks
Optimized primitives for collective multi-GPU communication
☆4,216Updated last week
Alternatives and similar repositories for nccl
Users that are interested in nccl are comparing it to the libraries listed below
Sorting:
- NCCL Tests☆1,334Updated last week
- Collective communications library with various primitives for multi-machine training.☆1,367Updated 3 weeks ago
- A fast GPU memory copy library based on NVIDIA GPUDirect RDMA technology☆1,264Updated 2 months ago
- CUDA Templates and Python DSLs for High-Performance Linear Algebra☆8,742Updated last 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…☆2,909Updated this week
- oneAPI Deep Neural Network Library (oneDNN)☆3,913Updated this week
- FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/☆1,471Updated last week
- Reference implementations of MLPerf® training benchmarks☆1,724Updated last week
- FlashInfer: Kernel Library for LLM Serving☆4,062Updated this week
- Source code examples from the Parallel Forall Blog☆1,311Updated last month
- Open MPI main development repository☆2,443Updated last week
- CUDA Library Samples☆2,164Updated last week
- [ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl☆1,801Updated 2 years ago
- A machine learning compiler for GPUs, CPUs, and ML accelerators☆3,670Updated this week
- Samples for CUDA Developers which demonstrates features in CUDA Toolkit☆8,435Updated 2 months ago
- Transformer related optimization, including BERT, GPT☆6,348Updated last year
- Reference implementations of MLPerf® inference benchmarks☆1,480Updated last week
- CUDA Core Compute Libraries☆2,019Updated this week
- Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training☆1,843Updated this week
- PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT☆2,884Updated this week
- Unified Communication X (mailing list - https://elist.ornl.gov/mailman/listinfo/ucx-group)☆1,502Updated this week
- ☆1,942Updated 2 years ago
- A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep lear…☆5,548Updated this week
- common in-memory tensor structure☆1,096Updated last month
- PyTorch extensions for high performance and large scale training.☆3,385Updated 6 months ago
- Low-precision matrix multiplication☆1,817Updated last year
- Tile primitives for speedy kernels☆2,873Updated last week
- A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.☆885Updated last week
- Examples demonstrating available options to program multiple GPUs in a single node or a cluster☆826Updated last month
- A unified library of state-of-the-art model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. …☆1,529Updated this week