SpatialUAV - Aerial-Aerial Collaboration Recognition: leaderboard
Cross-view matching between drones from SpatialUAV (Zhang et al., 2026): given one UAV view of a scene, the model selects the view taken by a second UAV of the same scene among clean candidate images. Accuracy in percent over 420 questions; higher is better.
Metric: Collaboration Recognition Accuracy (%; choose the UAV view that matches a given UAV view). Source: arxiv.org. Saturation forecast: Estimated already saturated. 18 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 92.6 | #121 |
| 2 | Claude Opus 4.7 | 85.7 | #75 |
| 3 | Qwen 3.6 27B | 61.9 | #339 |
| 4 | Qwen 3.5 9B | 37.9 | #803 |
| 5 | Qwen 3.6 35B A3B | 19.5 | #461 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=spatialuav-aerial-aerial-collaboration-recognition · How It Works · Data refreshed daily, snapshot 2026-11-12.