English

HeteGen: Heterogeneous Parallel Inference for Large Language Models on Resource-Constrained Devices

Performance 2024-03-05 v1 Distributed, Parallel, and Cluster Computing

Abstract

In recent times, the emergence of Large Language Models (LLMs) has resulted in increasingly larger model size, posing challenges for inference on low-resource devices. Prior approaches have explored offloading to facilitate low-memory inference but often suffer from efficiency due to I/O bottlenecks. To achieve low-latency LLMs inference on resource-constrained devices, we introduce HeteGen, a novel approach that presents a principled framework for heterogeneous parallel computing using CPUs and GPUs. Based on this framework, HeteGen further employs heterogeneous parallel computing and asynchronous overlap for LLMs to mitigate I/O bottlenecks. Our experiments demonstrate a substantial improvement in inference speed, surpassing state-of-the-art methods by over 317% at most.

Keywords

Cite

@article{arxiv.2403.01164,
  title  = {HeteGen: Heterogeneous Parallel Inference for Large Language Models on Resource-Constrained Devices},
  author = {Xuanlei Zhao and Bin Jia and Haotian Zhou and Ziming Liu and Shenggan Cheng and Yang You},
  journal= {arXiv preprint arXiv:2403.01164},
  year   = {2024}
}

Comments

MLSys 2024

R2 v1 2026-06-28T15:07:01.601Z