VB (Visibility Benchmark): leaderboard
Metric: FinalScore (0-1, shown times 100): 0.70 CAA + 0.15 MEFR + 0.10 SelRank + 0.05 ToMAcc over the 300 headline items (BASE, TEXT_FLIP and IMAGE_FLIP cells of 100 families); SelRank can be negative, so the score has no fixed lower bound; higher is better. Source: arxiv.org. Saturation forecast: Around June 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 72.8 | #333 |
| 2 | Gemini 3.1 Pro (Preview) | 72.7 | #54 |
| 3 | Gemini 2.5 Pro | 67.8 | #145 |
| 4 | GPT-5 | 62.5 | #91 |
| 5 | Claude Opus 4.5 | 57 | #79 |
| 6 | Gemma 3 12B | 50.5 | #666 |
| 7 | Claude 3.7 Sonnet | 47.6 | #241 |
| 8 | InternVL3-8B | 44.5 | #606 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=vb-visibility-benchmark · How It Works · Data refreshed daily, snapshot 2026-10-11.