cs.PF · 2026-08-31 · No. 101
Performance, 2026-08-31.
1 new papers in cs.PF. Titles, authors,
abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →
01 — The papers
1 entries-
01
Characterization of Request and Token Energy Costs for LLM Inference Workloads on GPU Platforms
Prabhu Vellaisamy, Vanessa Lam, Shawn Blanton, John Paul Shen
cs.PF · cs.DC · cs.LG
Large language model (LLM) inference serving is priced by tokens, but GPU energy is consumed over inference windows. This accounting mismatch makes token-normalized metrics incomplete, since average output-token energy can decrease even when total request energy increases. We characterize this behavior with a decomposed energy model: a fixed one-time prefill with a fixed generation setup cost, while each output-token generation step adds...
This edition is part of The Daily Abstract — cs.PF archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.
#D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.