KOREAN ASSOCIATION FOR ARTIFICIAL INTELLIGENCE AND LAW

KAAIL · AI LAW & POLICY RESEARCH

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

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

International ResearcharXivAbstract available

Preprint · Peer review status unconfirmed

Author

Chenxu Niu · Batuhan Sencer · Wei Zhang · Yong Chen

Publication DateLanguage en

Summary status

An abstract is provided, but an AI summary has not yet been generated.

Abstract Preview
Energy-efficient deployment of large language models requires understanding how model architecture, hardware, and servin…
Read more in the abstract and original text

Original text and source

Database
arXiv
Original paperRead more in the original text