高技术通讯2026，Vol.36Issue(7)：663-676,14.DOI:10.3772/j.issn.1002-0470.2026.07.001
大语言模型推理优化研究综述
A survey on optimization research for large language model inference
摘要
Abstract
The rapid growth in parameter size of large language model(LLM)and their autoregressive nature has led to a significant contradiction between high service costs for inference and low resource utilization efficiency.The per-formance of inference systems depends on various factors,including system architecture,algorithms,and models.The paper summarizes the challenges faced by inference systems from the request processing flow and model service architecture.It focuses on research advancements in request scheduling,resource management,token generation,distributed inference,and inference caching.The paper also analyzes key technologies for improving inference sys-tem efficiency.Finally,this work discusses the challenges of reducing inference service costs in heterogeneous hardware resources and text-to-image scenarios,suggesting that software-hardware co-design and hardware-aware strategies can further enhance resource efficiency and reduce inference costs.关键词
大模型服务/请求调度/缓存加速/显存管理/算子优化Key words
large language model serving/request scheduling/caching accelerate/memory management/op-erator optimization引用本文复制引用
胡存琛,孙凝晖,王卅..大语言模型推理优化研究综述[J].高技术通讯,2026,36(7):663-676,14.基金项目
国家自然科学基金(U24B6012)资助项目. （U24B6012）
