MuseBench - Static Visual Arts (Single-Select): leaderboard
Metric: Static-visual-arts chance-adjusted accuracy (%): raw single-select accuracy minus the per-item chance baseline, rescaled so random guessing scores 0 and a correct answer 1; a negative value is worse than chance; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 28 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 Plus | 64.36 |
| 2 | Claude Opus 4.6 | 58.51 |
| 3 | Qwen 3.5 397B A17B | 57.45 |
| 4 | GPT-5.4 | 54.24 |
| 5 | InternVL3-78B | 48.79 |
| 6 | Gemini 3.1 Pro (Preview) | 42.7 |
| 7 | Qwen2.5-Omni-7B | 40.76 |
| 8 | InternVL3-8B | 36.23 |
| 9 | Gemma 4 E4B | 32.55 |
| 10 | Kimi K2.5 | 24.05 |
| 11 | Grok 4.1 | 19.7 |
| 12 | GLM-4.5V | -2.97 |
Interactive version: theaggregate.ai/benchmark?slug=musebench-static-visual-arts-single-select · How It Works · Data refreshed daily, snapshot 2026-09-29.