Energy per Successful Goal: Goal-Level Energy Accounting for Agentic AI Systems
A new framework called A-LEMS has been introduced for measuring energy consumption in agentic AI systems. This framework shifts the focus from energy per inference to Energy per Successful Goal (EpG), which accounts for the total energy used across all attempts to achieve a goal. The findings indicate that agentic workflows consume significantly more energy than linear baselines, highlighting the importance of orchestration structure in energy costs.
- ▪A-LEMS redefines AI energy accounting from energy per inference to Energy per Successful Goal (EpG).
- ▪Agentic workflows consume 4.33 times higher mean energy per successful goal compared to linear baselines.
- ▪The Orchestration Overhead Index (OOI) measures the energy cost of orchestration relative to linear execution.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.22883 (cs) [Submitted on 20 May 2026] Title:Energy per Successful Goal: Goal-Level Energy Accounting for Agentic AI Systems Authors:Deepak Panigrahy, Aakash Tyagi View a PDF of the paper titled Energy per Successful Goal: Goal-Level Energy Accounting for Agentic AI Systems, by Deepak Panigrahy and 1 other authors View PDF HTML (experimental) Abstract:Current AI energy benchmarks measure consumption at the granularity of a single model invocation or training run. For classical single-turn workloads this unit remains coherent.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
