Specinfer: Accelerating Generative Large Language Model Serving With Tree-based Speculative Inference And Verification

Miao Xupeng, Oliaro Gabriele, Zhang Zhihao, Cheng Xinhao, Wang Zeyu, Zhang Zhengxin, Wong Rae Ying Yee, Zhu Alan, Yang Lijie, Shi Xiaoxiang, Shi Chunan, Chen Zhuoming, Arfeen Daiyaan, Abhyankar Reyna, Jia Zhihao. Arxiv 2023

[Paper] [Code]    
Has Code RAG

This paper introduces SpecInfer, a system that accelerates generative large language model (LLM) serving with tree-based speculative inference and verification. The key idea behind SpecInfer is leveraging small speculative models to predict the LLM’s outputs; the predictions are organized as a token tree, whose nodes each represent a candidate token sequence. The correctness of all candidate token sequences represented by a token tree is verified against the LLM in parallel using a novel tree-based parallel decoding mechanism. SpecInfer uses an LLM as a token tree verifier instead of an incremental decoder, which significantly reduces the end-to-end latency and computational requirement for serving generative LLMs while provably preserving model quality. Our evaluation shows that SpecInfer outperforms existing LLM serving systems by 1.5-2.8x for distributed LLM inference and by 2.6-3.5x for offloading-based LLM inference, while preserving the same generative performance. SpecInfer is publicly available at https://github.com/flexflow/FlexFlow/

Similar Work