KuangjuX / NVSHMEM-TutorialLinks
NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer
☆144Updated 2 months ago
Alternatives and similar repositories for NVSHMEM-Tutorial
Users that are interested in NVSHMEM-Tutorial are comparing it to the libraries listed below
Sorting:
- A lightweight design for computation-communication overlap.☆190Updated last month
- Tile-based language built for AI computation across all scales☆82Updated this week
- ☆112Updated 6 months ago
- We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel …☆190Updated 10 months ago
- DLSlime: Flexible & Efficient Heterogeneous Transfer Toolkit☆82Updated this week
- ☆102Updated last year
- DeeperGEMM: crazy optimized version☆73Updated 7 months ago
- ☆65Updated 7 months ago
- ☆66Updated 6 months ago
- High performance Transformer implementation in C++.☆142Updated 10 months ago
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆134Updated 6 months ago
- gLLM: Global Balanced Pipeline Parallelism System for Distributed LLM Serving with Token Throttling☆51Updated this week
- ☆34Updated last month
- Implement Flash Attention using Cute.☆97Updated 11 months ago
- A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.☆118Updated 6 months ago
- [HPCA 2026] A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆63Updated 2 weeks ago
- ☆90Updated 8 months ago
- ☆38Updated last month
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆96Updated 2 months ago
- Low overhead tracing library and trace visualizer for pipelined CUDA kernels☆116Updated last week
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆78Updated last year
- ☆148Updated 11 months ago
- Triton adapter for Ascend. Mirror of https://gitee.com/ascend/triton-ascend☆86Updated this week
- Multi-Level Triton Runner supporting Python, IR, PTX, and cubin.☆76Updated this week
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆102Updated 5 months ago
- ☆79Updated last month
- ☆324Updated 3 weeks ago
- Aims to implement dual-port and multi-qp solutions in deepEP ibrc transport☆68Updated 6 months ago
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆70Updated last week
- An experimental communicating attention kernel based on DeepEP.☆34Updated 4 months ago