CoChat
Sign Up
Cited by 121Natural Language Processing TechniquesTopic ModelingParallel Computing and Optimization Techniques

The DOI matches its Crossref record; 23 of the 24 references it cites were found in a registry or an index.

SpecInfer: Accelerating Large Language Model Serving with Tree-based Speculative Inference and Verification

Open access
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·2024·Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3
DOI 10.1145/3620666.3651335
Open this paper in CoChat — free account

Abstract

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 1.5-2.8× for distributed LLM inference and by 2.6-3.5× for offloading-based LLM inference, while preserving the same generative performance. SpecInfer is publicly available at https://github.com/flexflow/FlexFlow/

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1145/3620666.3651335

Cites

23 of 23 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 24 references

Research works printed but not linked: 29

  1. Charlie Chen, Sebastian Borgeaud, Geoffrey Irving, Jean-Baptiste Lespiau, Laurent Sifre, and John Jumper. Accelerating large language model decoding with speculative sampling. arXiv preprint arXiv:2302.01318, 2023. · arXiv
  2. Aakanksha Chowdhery, Sharan Narang, Jacob Devlin, Maarten Bosma, Gaurav Mishra, Adam Roberts, Paul Barham, Hyung Won Chung, Charles Sutton, Sebastian Gehrmann, et al. Palm: Scaling language modeling with pathways. arXiv preprint arXiv:2204.02311, 2022. · arXiv
  3. Tim Dettmers, Mike Lewis, Younes Belkada, and Luke Zettlemoyer. Gpt3. int8 (): 8-bit matrix multiplication for transformers at scale. Advances in Neural Information Processing Systems, 35:30318--30332, 2022.
  4. Elias Frantar, Saleh Ashkboos, Torsten Hoefler, and Dan Alistarh. Gptq: Accurate quantization for generative pre-trained transformers. In International Conference on Learning Representations, 2023.
  5. Freddy Gabbay and Avi Mendelson. Speculative execution based on value prediction. Citeseer, 1996.
  6. Joao Gante. Assisted generation: a new direction toward low-latency text generation, 2023.
  7. Sehoon Kim, Karttikeya Mangalam, Suhong Moon, John Canny, Jitendra Malik, Michael W. Mahoney, Amir Gholami, and Kurt Keutzer. Big little transformer decoder, 2023.
  8. Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Yu, Joseph E Gonzalez, Hao Zhang, and Ion Stoica. vllm: Easy, fast, and cheap llm serving with pagedattention. See https://vllm.ai/ (accessed 9 August 2023), 2023.
  9. Yaniv Leviathan, Matan Kalman, and Yossi Matias. Fast inference from transformers via speculative decoding. arXiv preprint arXiv:2211.17192, 2022. · arXiv
  10. Jiachang Liu, Dinghan Shen, Yizhe Zhang, Bill Dolan, Lawrence Carin, and Weizhu Chen. What makes good in-context examples for gpt-3? arXiv preprint arXiv:2101.06804, 2021. · arXiv
  11. Xupeng Miao, Gabriele Oliaro, Zhihao Zhang, Xinhao Cheng, Hongyi Jin, Tianqi Chen, and Zhihao Jia. Towards efficient generative large language model serving: A survey from algorithms to systems. arXiv preprint arXiv:2312.15234, 2023. · arXiv
  12. Xupeng Miao, Gabriele Oliaro, Zhihao Zhang, Xinhao Cheng, Zeyu Wang, Rae Ying Yee Wong, Zhuoming Chen, Daiyaan Arfeen, Reyna Abhyankar, and Zhihao Jia. Specinfer: Accelerating generative llm serving with speculative inference and token tree verification. arXiv preprint arXiv:2305.09781, 2023. · arXiv
  13. Xupeng Miao, Chunan Shi, Jiangfei Duan, Xiaoli Xi, Dahua Lin, Bin Cui, and Zhihao Jia. Spotserve: Serving generative large language models on preemptible instances. arXiv preprint arXiv:2311.15566, 2023. · arXiv
  14. OpenAI. Gpt-4 technical report, 2023.
  15. Gunho Park, Baeseong Park, Se Jung Kwon, Byeongwook Kim, Youngjoo Lee, and Dongsoo Lee. nuqmm: Quantized matmul for efficient inference of large-scale generative language models. arXiv preprint arXiv:2206.09557, 2022. · arXiv
  16. Baolin Peng, Chunyuan Li, Pengcheng He, Michel Galley, and Jianfeng Gao. Instruction tuning with gpt-4. arXiv preprint arXiv:2304.03277, 2023. · arXiv
  17. Jack W Rae, Sebastian Borgeaud, Trevor Cai, Katie Millican, Jordan Hoffmann, Francis Song, John Aslanides, Sarah Henderson, Roman Ring, Susannah Young, et al. Scaling language models: Methods, analysis & insights from training gopher. arXiv preprint arXiv:2112.11446, 2021. · arXiv
  18. Teven Le Scao, Angela Fan, Christopher Akiki, Ellie Pavlick, Suzana Ilić, Daniel Hesslow, Roman Castagné, Alexandra Sasha Luccioni, François Yvon, Matthias Gallé, et al. Bloom: A 176b-parameter open-access multilingual language model. arXiv preprint arXiv:2211.05100, 2022. · arXiv
  19. Ying Sheng, Lianmin Zheng, Binhang Yuan, Zhuohan Li, Max Ryabinin, Daniel Y. Fu, Zhiqiang Xie, Beidi Chen, Clark Barrett, Joseph E. Gonzalez, Percy Liang, Christopher Ré, Ion Stoica, and Ce Zhang. High-throughput generative inference of large language models with a single gpu, 2023.
  20. James E Smith. A study of branch prediction strategies. In 25 years of the international symposia on Computer architecture (selected papers), pages 202--215, 1998.
  21. Shaden Smith, Mostofa Patwary, Brandon Norick, Patrick LeGresley, Samyam Rajbhandari, Jared Casper, Zhun Liu, Shrimai Prabhumoye, George Zerveas, Vijay Korthikanti, et al. Using deepspeed and mega-tron to train megatron-turing nlg 530b, a large-scale generative language model. arXiv preprint arXiv:2201.11990, 2022. · arXiv
  22. Colin Unger, Zhihao Jia, Wei Wu, Sina Lin, Mandeep Baines, Carlos Efrain Quintero Narvaez, Vinay Ramakrishnaiah, Nirmal Prajapati, Pat McCormick, Jamaludin Mohd-Yusof, Xi Luo, Dheevatsa Mudigere, Jongsoo Park, Misha Smelyanskiy, and Alex Aiken. Unity: Accelerating DNN training through joint optimization of algebraic transformations and parallelization. In 16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22), pages 267--284, Carlsbad, CA, July 2022. USENIX Association.
  23. Guangxuan Xiao, Ji Lin, Mickael Seznec, Julien Demouth, and Song Han. Smoothquant: Accurate and efficient post-training quantization for large language models. arXiv preprint arXiv:2211.10438, 2022. · arXiv
  24. Nan Yang, Tao Ge, Liang Wang, Binxing Jiao, Daxin Jiang, Linjun Yang, Rangan Majumder, and Furu Wei. Inference with reference: Lossless acceleration of large language models. arXiv preprint arXiv:2304.04487, 2023. · arXiv
  25. Zhewei Yao, Reza Yazdani Aminabadi, Minjia Zhang, Xiaoxia Wu, Conglong Li, and Yuxiong He. Zeroquant: Efficient and affordable post-training quantization for large-scale transformers. Advances in Neural Information Processing Systems, 35:27168--27183, 2022.
  26. Gyeong-In Yu, Joo Seong Jeong, Geon-Woo Kim, Soojeong Kim, and Byung-Gon Chun. Orca: A distributed serving system for Transformer-Based generative models. In 16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22), pages 521--538, Carlsbad, CA, July 2022. USENIX Association.
  27. Haoyu Zhang, Jianjun Xu, and Ji Wang. Pretraining-based natural language generation for text summarization. arXiv preprint arXiv:1902.09243, 2019. · arXiv
  28. Susan Zhang, Stephen Roller, Naman Goyal, Mikel Artetxe, Moya Chen, Shuohui Chen, Christopher Dewan, Mona Diab, Xian Li, Xi Victoria Lin, et al. Opt: Open pre-trained transformer language models. arXiv preprint arXiv:2205.01068, 2022. · arXiv
  29. Lianmin Zheng Chengfan Jia Minmin Sun Zhao Wu Cody Hao Yu Ameer Haj-Ali Yida Wang Jun Yang Danyang Zhuo Koushik Sen et al. Ansor: Generating high-performance tensor programs for deep learning. In 14th {USENIX} Symposium on Operating Systems Design and Implementation ({OSDI} 20) pages 863--879 2020.

Cited by

Sorted by citations

See all 121

Related

Metadata

Authors
Miao, X., Oliaro, G., Zhang, Z., Cheng, X., Wang, Z., Zhang, Z., Wong, R. Y. Y., Zhu, A., Yang, L., Shi, X., Shi, C., Chen, Z., Arfeen, D., Abhyankar, R., & Jia, Z.
Affiliations
Shanghai Jiao Tong University (CN)
ORCID
Venue
Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3
Publisher
ACM
Published
2024-04-27
Pages
932-949
DOI
10.1145/3620666.3651335
OpenAlex
W4395112660
Cited by
114
References
24
Citations by year
2026: 54 · 2025: 50 · 2024: 9
FWCI
33.35
Type
conference paper
Language
en
Open access
Licence
Locations
  • International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion, cc-by)
  • arXiv (Cornell University) (cc-by)
Funders
  • NSF (CNS-2147909, CNS-2211882, CNS-2239351)
  • Amazon Research Award
  • Cisco Research Award
  • Google Faculty Research Award
  • Meta Research Award
  • Oracle Research Award
  • Qualcomm Innovation Fellowship
  • Samsung GRO Research Award
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
  • Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
  • Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
Keywords
large language model serving, parallel decoding, inference latency reduction, distributed inference, inference offloading

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review