paper-with-me

Papers

HOBBIT: A Mixed Precision Expert Offloading System for Fast MoE Inference

2024-11-03 · Peng Tang, Jiacheng Liu, Xiaofeng Hou, YiFei PU, Jing Wang, Pheng-Ann Heng, Chao Li, Minyi Guo

The Mixture-of-Experts (MoE) architecture has demonstrated significant advantages in the era of Large Language Models (LLMs), offering enhanced capabilities with reduced inference costs. However, deploying MoE-based LLMs on memoryconstrained edge devices remains challenging due to their substantial memory requirements. While existing expertoffloading methods alleviate the memory requirements, they often incur significant expert-loading costs or compromise model accuracy. We present HOBBIT, a mixed precision expert offloading system to enable flexible and efficient MoE inference. Our key insight is that dynamically replacing less critical cache-miss experts with low precision versions can substantially reduce expert-loading latency while preserving model accuracy. HOBBIT introduces three innovative techniques that map the natural hierarchy of MoE computation: (1) a token-level dynamic expert loading mechanism, (2) a layer-level adaptive expert prefetching technique, and (3) a sequence-level multidimensional expert caching policy. These innovations fully leverage the benefits of mixedprecision expert inference. By implementing HOBBIT on top of the renowned LLM inference framework Llama.cpp, we evaluate its performance across different edge devices with representative MoE models. The results demonstrate that HOBBIT achieves up to a 9.93x speedup in decoding compared to state-of-the-art MoE offloading systems.

📄 PDF Abstract BibTeX arXiv:2411.01433

Code (0)

등록된 구현이 없습니다.

Tasks

Mixture-of-Experts

Methods 이 논문이 사용한 방법론

MoE 설명 없음

Similar Papers 제목 키워드 기반

Dynamic Expert Quantization for Scalable Mixture-of-Experts Inference

2025-11-19 · Kexin Chu, Dawei Xiang, Zixu Shen, Yiwei Yang 외 arxiv

Mixture-of-Experts (MoE) has become a practical architecture for scaling LLM capacity while keeping per-token compute modest, but deploying MoE models on a single, memory-limited GPU remains difficult because expert weig…

DyMoE: Dynamic Expert Orchestration with Mixed-Precision Quantization for Efficient MoE Inference on Edge

2026-03-19 · Yuegui Huang, Zhiyuan Fang, Weiqi Luo, Ruoyu Wu 외 arxiv

Despite the computational efficiency of MoE models, the excessive memory footprint and I/O overhead inherent in multi-expert architectures pose formidable challenges for real-time inference on resource-constrained edge p…

Computational Efficiency

Context-Aware Mixture-of-Experts Inference on CXL-Enabled GPU-NDP Systems

2025-12-04 · Zehao Fan, Zhenyu Liu, Yunzhen Liu, Yayue Hou 외 arxiv

Mixture-of-Experts (MoE) models scale large language models through conditional computation, but inference becomes memory-bound once expert weights exceed the capacity of GPU memory. In this case, weights must be offload…

Bandwidth-Efficient Adaptive Mixture-of-Experts via Low-Rank Compensation

2025-12-18 · Zhenyu Liu, Yunzhen Liu, Zehao Fan, Garrett Gagnon 외 arxiv

Mixture-of-Experts (MoE) models scale capacity via sparse activation but stress memory and bandwidth. Offloading alleviates GPU memory by fetching experts on demand, yet token-level routing causes irregular transfers tha…

Fast Inference of Mixture-of-Experts Language Models with Offloading

2023-12-28 · Artyom Eliseev, Denis Mazur

With the widespread adoption of Large Language Models (LLMs), many deep learning practitioners are looking for strategies of running these models more efficiently. One such strategy is to use sparse Mixture-of-Experts (M…

Mixture-of-ExpertsQuantization