microsoft / tokenweaveLinks
Efficient Compute-Communication Overlap for Distributed LLM Inference
☆63Updated last month
Alternatives and similar repositories for tokenweave
Users that are interested in tokenweave are comparing it to the libraries listed below
Sorting:
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆146Updated 2 months ago
- ☆46Updated 7 months ago
- Tile-based language built for AI computation across all scales☆85Updated this week
- Nex Venus Communication Library☆64Updated 3 weeks ago
- ☆79Updated last month
- A lightweight design for computation-communication overlap.☆194Updated 2 months ago
- DLSlime: Flexible & Efficient Heterogeneous Transfer Toolkit☆82Updated this week
- gLLM: Global Balanced Pipeline Parallelism System for Distributed LLM Serving with Token Throttling☆51Updated last week
- DeeperGEMM: crazy optimized version☆73Updated 7 months ago
- Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding☆36Updated last week
- Stateful LLM Serving☆89Updated 9 months ago
- SpInfer: Leveraging Low-Level Sparsity for Efficient Large Language Model Inference on GPUs☆59Updated 8 months ago
- ☆65Updated 7 months ago
- Microsoft Collective Communication Library☆66Updated last year
- Aims to implement dual-port and multi-qp solutions in deepEP ibrc transport☆70Updated 7 months ago
- ☆47Updated 11 months ago
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆70Updated 2 weeks ago
- A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.☆118Updated 6 months ago
- [NeurIPS 2025] ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive☆50Updated 2 months ago
- Framework to reduce autotune overhead to zero for well known deployments.☆90Updated 2 months ago
- ☆67Updated 6 months ago
- ☆52Updated 6 months ago
- GPTQ inference TVM kernel☆40Updated last year
- Debug print operator for cudagraph debugging☆14Updated last year
- ☆60Updated 3 weeks ago
- [DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"☆94Updated 6 months ago
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆200Updated last year
- Multi-Level Triton Runner supporting Python, IR, PTX, and cubin.☆76Updated last week
- FractalTensor is a programming framework that introduces a novel approach to organizing data in deep neural networks (DNNs) as a list of …☆31Updated 11 months ago
- ☆19Updated last year