FlexFusion / FlexFusionLinks
The official implementation for the intra-stage fusion technique introduced in https://arxiv.org/abs/2409.13221
☆26Updated 6 months ago
Alternatives and similar repositories for FlexFusion
Users that are interested in FlexFusion are comparing it to the libraries listed below
Sorting:
- [NeurIPS 2024] Efficient LLM Scheduling by Learning to Rank☆62Updated last year
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆142Updated last month
- gLLM: Global Balanced Pipeline Parallelism System for Distributed LLM Serving with Token Throttling☆42Updated last month
- ☆75Updated 3 weeks ago
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆189Updated last year
- A lightweight design for computation-communication overlap.☆183Updated 3 weeks ago
- ☆124Updated 11 months ago
- Scalable long-context LLM decoding that leverages sparsity—by treating the KV cache as a vector storage system.☆98Updated last month
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆65Updated 2 weeks ago
- nnScaler: Compiling DNN models for Parallel Training☆118Updated last month
- DLSlime: Flexible & Efficient Heterogeneous Transfer Toolkit☆75Updated this week
- [DAC'25] Official implement of "HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference"☆85Updated 4 months ago
- Stateful LLM Serving☆88Updated 7 months ago
- ☆27Updated 7 months ago
- ☆30Updated 3 weeks ago
- [NeurIPS 2025] ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive☆47Updated last month
- Artifact for "Marconi: Prefix Caching for the Era of Hybrid LLMs" [MLSys '25 Outstanding Paper Award, Honorable Mention]☆42Updated 8 months ago
- Tile-based language built for AI computation across all scales☆74Updated this week
- ☆43Updated 6 months ago
- A framework for generating realistic LLM serving workloads☆76Updated last month
- High performance Transformer implementation in C++.☆140Updated 9 months ago
- [EuroSys'25] Mist: Efficient Distributed Training of Large Language Models via Memory-Parallelism Co-Optimization☆21Updated 3 months ago
- ☆79Updated 6 months ago
- Efficient Compute-Communication Overlap for Distributed LLM Inference☆62Updated last week
- ☆32Updated last week
- ☆57Updated last week
- ArkVale: Efficient Generative LLM Inference with Recallable Key-Value Eviction (NIPS'24)☆43Updated 10 months ago
- ☆64Updated 6 months ago
- NEO is a LLM inference engine built to save the GPU memory crisis by CPU offloading☆67Updated 4 months ago
- Allow torch tensor memory to be released and resumed later☆160Updated last week