# Preprint finds AI code optimization usually increases execution energy

_Published Monday, October 26, 2026 at 9:08 AM EDT · AI, Science · Latest · Tier 2 — Notable_

A preprint testing 11 open-source models on all 1,000 EffiBench problems found that 42 of 44 configurations produced code consuming more energy per problem than canonical human reference solutions. The excess ranged from +4% to +886%.

Researchers measured energy during iterative model refinement and code execution. Limiting maximum output tokens reduced inference energy by up to 69% against the default configuration. Mixture-of-Experts models used substantially less inference energy than equivalently-sized dense models without sacrificing correctness.

The two configurations that improved execution energy achieved only marginal gains, requiring on the order of 10^6 to 10^7 executions to offset optimization energy.

## Sources

- [cs.AI updates on arXiv.org](https://arxiv.org/abs/2610.32891)

---
Canonical: https://techandbusiness.org/newswire/6jjabxs0QX-dvEu2hpjUIt
Published: 2026-10-26T13:08:15.049Z
Story chronology: 2026-10-26T04:00:00.000Z
Retrieved: 2026-10-26T15:59:35.552Z
Publisher: Tech & Business (techandbusiness.org)