VisReason (Vision-Centric) - Spot the Difference: leaderboard
VisReason Spot the Difference is a perceptual-reasoning category of VisReason (2026): the model must find every difference between two scenes and localize each one with a bounding box. The score is accuracy in percent, higher is better.
Metric: Accuracy (%) on the Spot the Difference category (perceptual reasoning: find every difference between two scenes and localize each one with a bounding box) of VisReason, answers scored by format (regular-expression match, GPT-5-mini judge for open-ended answers, IoU above 0.5 for boxes), zero-shot with chain-of-thought prompt templates, one per answer format; rows not shaded gray in the paper use explicit reasoning (GPT-5 and Gemini 3 Pro at low reasoning effort); higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 20 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | Low | 4.8 | #119 |
| 2 | GPT-5.2 | Thinking | 4.7 | #185 |
| 3 | GPT-5 | Low | 2.8 | #178 |
| 4 | O4 Mini | 2.7 | #368 | |
| 5 | GPT-5 Mini | Low | 2.2 | #364 |
| 6 | Qwen 3 VL 235B A22B (Thinking) | 1.6 | #459 | |
| 7 | Qwen 2.5 VL 32B Instruct | 1.5 | #1062 | |
| 8 | Qwen 3 VL 32B (Thinking) | 1.3 | #573 | |
| 9 | Qwen 3 VL 8B Instruct | 1.1 | #1029 | |
| 10 | GPT-4o | 0.8 | #751 | |
| 11 | Keye-VL-1.5-8B | Thinking | 0.8 | #1720 |
| 12 | Qwen 3 VL 8B (Thinking) | 0.7 | #910 | |
| 13 | MiMo-VL-7B-RL | 0.7 | #1155 | |
| 14 | Gemini 2.5 Flash | 0.5 | #463 | |
| 15 | Qwen 3 VL 30B A3B (Thinking) | 0.5 | #719 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=visreason-vision-centric-spot-the-difference · How It Works · Data refreshed daily, snapshot 2026-11-21.