ArtECulture - English: leaderboard
Metric: Accuracy (%; English-speaking culture labels; zero-shot prediction of the majority emotion (nine categories) that the target culture perceives in each of the 679 test artworks; original models without the authors retrieval knowledge or fine-tuning). Source: arxiv.org. Saturation forecast: Around 2031. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 4 31B (IT) | 59.35 |
| 2 | Claude Opus 4.8 | 59.2 |
| 3 | Gemini 3.5 Flash | 59.2 |
| 4 | Claude Sonnet 4.6 | 58.62 |
| 5 | GPT-5.5 | 58.62 |
| 6 | Gemini 3 Flash | 58.62 |
| 7 | Qwen 3.5 27B | 58.32 |
| 8 | Qwen 3.6 27B | 57.58 |
| 9 | Qwen 3.6 35B A3B | 55.38 |
| 10 | Qwen 3.5 9B | 55.23 |
| 11 | aya-vision-8B | 54.34 |
| 12 | Gemma 3 12B (IT) | 53.76 |
| 13 | Pixtral-12B | 52.72 |
| 14 | Qwen 3 VL 8B | 50.37 |
Interactive version: theaggregate.ai/benchmark?slug=arteculture-english · How It Works · Data refreshed daily, snapshot 2026-09-26.