AesGuide - Expert Assessment: leaderboard

Metric: Expert assessment score on 100 sampled AesGuide photos: ten photography professionals score each critique 0, 1 or 2 with the judge's prompts, mean of completeness, preciseness and relevance on a 0 to 2 scale; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 10 models tracked.

Top models

#ModelScoreOverall rank
1GPT-4o1.15#333
2MiniCPM-V-2.60.92#825

No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.

Interactive version: theaggregate.ai/benchmark?slug=aesguide-expert-assessment · How It Works · Data refreshed daily, snapshot 2026-10-11.