The Token-Energy Law of LLM Decode: Active Bytes and Realized Time Account for Joules per Token

Paper recorded by Signals 4 on 2026-10-20 in cs.LG. Abstract reproduced from arXiv; link to the original below.

Published 2026-10-20 on arXiv · recorded by Signals 4 on 2026-10-21

Category: cs.LG · 机器学习 · first seen 2026-10-21

Abstract

Energy-efficient deployment of large language models requires understanding how model architecture, hardware, and serving settings shape energy per generated token. We introduce the Token-Energy Law, a compact, interpretable model of GPU-board energy per output token for dense, full-history decoding. It relates token energy to two core dimensions: active bytes and realized decode time. Active byte

Read on arXiv →