Benchmarking the Energy Savings with Speculative Decoding Strategies

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Dutta, Rohit, Koley, Paramita, Poddar, Soham, Misra, Janardan, Podder, Sanjay, Balani, Naveen, Ghosh, Saptarshi, Ganguly, Niloy
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908823505076224
author Dutta, Rohit
Koley, Paramita
Poddar, Soham
Misra, Janardan
Podder, Sanjay
Balani, Naveen
Ghosh, Saptarshi
Ganguly, Niloy
author_facet Dutta, Rohit
Koley, Paramita
Poddar, Soham
Misra, Janardan
Podder, Sanjay
Balani, Naveen
Ghosh, Saptarshi
Ganguly, Niloy
contents Speculative decoding has emerged as an effective method to reduce latency and inference cost of LLM inferences. However, there has been inadequate attention towards the energy requirements of these models. To address this gap, this paper presents a comprehensive survey of energy requirements of speculative decoding strategies, with detailed analysis on how various factors -- model size and family, speculative decoding strategies, and dataset characteristics -- influence the energy optimizations.
format Preprint
id arxiv_https___arxiv_org_abs_2602_09113
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Benchmarking the Energy Savings with Speculative Decoding Strategies
Dutta, Rohit
Koley, Paramita
Poddar, Soham
Misra, Janardan
Podder, Sanjay
Balani, Naveen
Ghosh, Saptarshi
Ganguly, Niloy
Machine Learning
Artificial Intelligence
Computation and Language
Speculative decoding has emerged as an effective method to reduce latency and inference cost of LLM inferences. However, there has been inadequate attention towards the energy requirements of these models. To address this gap, this paper presents a comprehensive survey of energy requirements of speculative decoding strategies, with detailed analysis on how various factors -- model size and family, speculative decoding strategies, and dataset characteristics -- influence the energy optimizations.
title Benchmarking the Energy Savings with Speculative Decoding Strategies
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2602.09113