llm-energy-per-prompt-image-vs-text-60x

OUT premise — summaries/2026/08/24/wiki-Large_language_model-chunk-3.md

Created 2026-08-24T17:11:16+00:00

Per Luccioni et al. (ACM FAccT 2024), a single image generation prompt requires ~2.91 Wh (up to 11.49 Wh for the least-efficient model) versus ~0.05 Wh for text generation/summarization, a roughly 60x energy difference