PuMVR: leaderboard
Metric: Script Consistency Rate (%; instances answered correctly in all three scripts, zero-shot). Source: arxiv.org. Saturation forecast: Around December 2026. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Flash | 85.6 |
| 2 | Grok 4.1 Fast (Reasoning) | 79.73 |
| 3 | GPT-4o | 78.13 |
| 4 | Claude Sonnet 4 | 77.09 |
| 5 | Qwen 2 VL 72B Instruct | 54.67 |
| 6 | Qwen 2 VL 7B Instruct | 41.33 |
| 7 | Llama 3.2 11B Instruct | 27.47 |
Interactive version: theaggregate.ai/benchmark?slug=pumvr · How It Works · Data refreshed daily, snapshot 2026-09-25.