microsoft / tokenweaveLinks
Efficient Compute-Communication Overlap for Distributed LLM Inference
☆66Updated 2 months ago
Alternatives and similar repositories for tokenweave
Users that are interested in tokenweave are comparing it to the libraries listed below
Sorting:
- ☆47Updated 8 months ago
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆150Updated 3 months ago
- Stateful LLM Serving☆92Updated 9 months ago
- A lightweight design for computation-communication overlap.☆206Updated last week
- ☆80Updated 2 months ago
- DeeperGEMM: crazy optimized version☆74Updated 7 months ago
- DLSlime: Flexible & Efficient Heterogeneous Transfer Toolkit☆89Updated this week
- ☆47Updated last year
- ☆65Updated 8 months ago
- Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding☆70Updated last month
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆80Updated last week
- Tile-based language built for AI computation across all scales☆109Updated last week
- Nex Venus Communication Library☆68Updated last month
- Microsoft Collective Communication Library☆66Updated last year
- Thunder Research Group's Collective Communication Library☆44Updated 5 months ago
- Aims to implement dual-port and multi-qp solutions in deepEP ibrc transport☆73Updated 7 months ago
- ☆67Updated this week
- A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.☆123Updated last week
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆66Updated 3 months ago
- ☆73Updated 11 months ago
- [NeurIPS 2025] ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive☆58Updated 3 weeks ago
- A NCCL extension library, designed to efficiently offload GPU memory allocated by the NCCL communication library.☆74Updated 2 weeks ago
- ☆92Updated 8 months ago
- SpInfer: Leveraging Low-Level Sparsity for Efficient Large Language Model Inference on GPUs☆59Updated 9 months ago
- ☆19Updated last year
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆104Updated 6 months ago
- ☆36Updated 2 months ago
- Framework to reduce autotune overhead to zero for well known deployments.☆91Updated 3 months ago
- ☆77Updated 7 months ago
- ☆144Updated last year