sii-research / VCCLLinks
Venus Collective Communication Library, supported by SII and Infrawaves.
☆95Updated this week
Alternatives and similar repositories for VCCL
Users that are interested in VCCL are comparing it to the libraries listed below
Sorting:
- DLSlime: Flexible & Efficient Heterogeneous Transfer Toolkit☆67Updated this week
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆135Updated 3 weeks ago
- A lightweight design for computation-communication overlap.☆179Updated 3 weeks ago
- A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.☆111Updated 4 months ago
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆119Updated 5 months ago
- ☆91Updated last week
- DeeperGEMM: crazy optimized version☆71Updated 5 months ago
- Virtualized Elastic KV Cache for Dynamic GPU Sharing and Beyond☆99Updated last week
- ☆85Updated 6 months ago
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆220Updated 2 months ago
- Tile-based language built for AI computation across all scales☆66Updated last week
- ☆300Updated last week
- Fast and memory-efficient exact attention☆94Updated this week
- A Triton JIT runtime and ffi provider in C++☆25Updated this week
- We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel …☆186Updated 8 months ago
- Compare different hardware platforms via the Roofline Model for LLM inference tasks.☆115Updated last year
- High performance Transformer implementation in C++.☆135Updated 8 months ago
- DeepSeek-V3/R1 inference performance simulator☆169Updated 6 months ago
- Triton adapter for Ascend. Mirror of https://gitee.com/ascend/triton-ascend☆76Updated 2 weeks ago
- [DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"☆72Updated 4 months ago
- ☆46Updated 9 months ago
- ☆64Updated 5 months ago
- ☆99Updated 4 months ago
- ☆42Updated 5 months ago
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆60Updated this week
- ☆99Updated last year
- Aims to implement dual-port and multi-qp solutions in deepEP ibrc transport☆63Updated 5 months ago
- Efficient Compute-Communication Overlap for Distributed LLM Inference☆58Updated last week
- gLLM: Global Balanced Pipeline Parallelism System for Distributed LLM Serving with Token Throttling☆41Updated last week
- ☆95Updated 6 months ago