DiCoBench - Spatial Commonality: leaderboard
Metric: Accuracy (%) on the spatial-commonality task (locate the shared element on a 6 x 6 grid) of DiCoBench (765 multi-image samples at near-2K resolution): 5-option multiple choice (four candidate image pairs or instructions plus a no-visible-commonality option, chance 20%), answer letter matched exactly, VLMEvalKit settings at temperature 0; higher is better. Source: arxiv.org. Saturation forecast: Around August 2027. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 58.4 |
| 2 | Qwen 3.5 35B A3B | 49.6 |
| 3 | O4 Mini | 48 |
| 4 | Gemini 3 Flash | 41.7 |
| 5 | Gemma 3 27B (IT) | 36.8 |
| 6 | GPT-5 | 36 |
| 7 | Gemma 3 12B (IT) | 29.6 |
| 8 | GPT-4.1 | 27.2 |
| 9 | GPT-4.1 Mini | 25.6 |
| 10 | Qwen 2.5 VL 7B Instruct | 23.2 |
| 11 | Qwen 2.5 VL 32B Instruct | 21.6 |
Interactive version: theaggregate.ai/benchmark?slug=dicobench-spatial-commonality · How It Works · Data refreshed daily, snapshot 2026-09-29.