MuseBench - Cinematic Arts (Single-Select): leaderboard
Metric: Cinematic-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 | 68.88 |
| 2 | Claude Opus 4.6 | 63.26 |
| 3 | Qwen 3.5 397B A17B | 62.65 |
| 4 | GPT-5.4 | 56.5 |
| 5 | InternVL3-78B | 47.98 |
| 6 | InternVL3-8B | 43.41 |
| 7 | Gemini 3.1 Pro (Preview) | 43.16 |
| 8 | Gemma 4 E4B | 39.4 |
| 9 | Qwen2.5-Omni-7B | 36.47 |
| 10 | Kimi K2.5 | 23.06 |
| 11 | GLM-4.5V | 16.17 |
| 12 | Grok 4.1 | 14.19 |
Interactive version: theaggregate.ai/benchmark?slug=musebench-cinematic-arts-single-select · How It Works · Data refreshed daily, snapshot 2026-09-29.