ArtECulture - Chinese: leaderboard
Metric: Accuracy (%; Chinese-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 2030. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.8 | 51.33 |
| 2 | Gemini 3 Flash | 51.25 |
| 3 | Gemini 3.5 Flash | 50.81 |
| 4 | Gemma 4 31B (IT) | 46.24 |
| 5 | Qwen 3.5 27B | 45.36 |
| 6 | Claude Sonnet 4.6 | 45.21 |
| 7 | Qwen 3.5 9B | 43.89 |
| 8 | Qwen 3.6 35B A3B | 43 |
| 9 | GPT-5.5 | 41.83 |
| 10 | Qwen 3 VL 8B | 41.38 |
| 11 | Qwen 3.6 27B | 38.73 |
| 12 | Gemma 3 12B (IT) | 36.23 |
| 13 | aya-vision-8B | 33.87 |
| 14 | Pixtral-12B | 17.08 |
Interactive version: theaggregate.ai/benchmark?slug=arteculture-chinese · How It Works · Data refreshed daily, snapshot 2026-09-26.