NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process communication and coordination overheads by allowing programmers to perform one-sided communication from within CUDA kernels and on CUDA streams.
☆482Mar 10, 2026Updated last week
Alternatives and similar repositories for nvshmem
Users that are interested in nvshmem are comparing it to the libraries listed below
Sorting:
- ☆32Jul 2, 2025Updated 8 months ago
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆170Feb 11, 2026Updated last month
- Low overhead tracing library and trace visualizer for pipelined CUDA kernels☆134Nov 26, 2025Updated 3 months ago
- Distributed Compiler based on Triton for Parallel Systems☆1,386Mar 11, 2026Updated last week
- ☆90May 31, 2025Updated 9 months ago
- A lightweight design for computation-communication overlap.☆225Jan 20, 2026Updated 2 months ago
- An experimental communicating attention kernel based on DeepEP.☆35Jul 29, 2025Updated 7 months ago
- ☆65Apr 26, 2025Updated 10 months ago
- Perplexity GPU Kernels☆564Nov 7, 2025Updated 4 months ago
- A NCCL extension library, designed to efficiently offload GPU memory allocated by the NCCL communication library.☆99Dec 17, 2025Updated 3 months ago
- DeeperGEMM: crazy optimized version☆75May 5, 2025Updated 10 months ago
- ☆352Jan 28, 2026Updated last month
- High-performance LLM operator library built on TileLang.☆93Updated this week
- ☆53Feb 24, 2026Updated 3 weeks ago
- [DEPRECATED] Moved to ROCm/rocm-systems repo☆146Mar 10, 2026Updated last week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,273Aug 28, 2025Updated 6 months ago
- ☆38Aug 7, 2025Updated 7 months ago
- Tutorial Exercises and Code for GPU Communications Tutorial at HOT Interconnects 2025☆31Oct 22, 2025Updated 4 months ago
- Tile-based language built for AI computation across all scales☆138Updated this week
- A Quirky Assortment of CuTe Kernels☆861Updated this week
- DLSlime: Flexible & Efficient Heterogeneous Transfer Toolkit☆91Jan 26, 2026Updated last month
- Nex Venus Communication Library☆74Nov 17, 2025Updated 4 months ago
- ☆52May 19, 2025Updated 10 months ago
- Tilus is a tile-level kernel programming language with explicit control over shared memory and registers.☆456Updated this week
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆490Updated this week
- A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.☆803Updated this week
- DeepSeek-V3.2-Exp DSA Warmup Lightning Indexer training operator based on tilelang☆44Nov 19, 2025Updated 4 months ago
- Noisy language compiler☆17Jul 31, 2024Updated last year
- We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel …☆192Jan 28, 2025Updated last year
- Optimized primitives for collective multi-GPU communication☆4,531Updated this week
- [Archived] For the latest updates and community contribution, please visit: https://github.com/Ascend/TransferQueue or https://gitcode.co…☆13Jan 16, 2026Updated 2 months ago
- incubator repo for CUDA-TileIR backend☆120Mar 14, 2026Updated last week
- NVIDIA Inference Xfer Library (NIXL)☆945Updated this week
- a simple API to use CUPTI☆10Aug 19, 2025Updated 7 months ago
- A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training☆676Updated this week
- Allow torch tensor memory to be released and resumed later☆225Mar 10, 2026Updated last week
- ☆139Aug 18, 2025Updated 7 months ago
- Fastest kernels written from scratch☆559Sep 18, 2025Updated 6 months ago
- ☆36Mar 7, 2025Updated last year