CVSBench - FOV Satellite-to-Ground: leaderboard
CVSBench (2026) tests cross-view spatial reasoning with paired street-view and satellite images: models infer attributes of the unseen view and localize the street-view camera on the satellite image, on four-choice and binary questions. This board is accuracy in percent on FOV-subset questions that infer limited field-of-view street-view attributes from a satellite image; higher is better.
Metric: Accuracy (%) on FOV-subset satellite-to-ground questions (limited field-of-view street images); four-choice and binary-choice questions over paired street-view and satellite images (Gemini-2.5-flash-generated, human-verified), test split; higher is better. Source: arxiv.org. Saturation forecast: Around 2031. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 Chat | 54.1 | #444 |
| 2 | Gemma 3 4B (IT) | 47.99 | #3831 |
| 3 | GPT-4o | 46.8 | #741 |
| 4 | Qwen 3 VL 8B Instruct | 46.7 | #981 |
| 5 | Qwen 3 VL 4B Instruct | 44.2 | #1279 |
| 6 | DeepSeek V3.2 | 38.68 | #406 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=cvsbench-fov-satellite-to-ground · How It Works · Data refreshed daily, snapshot 2026-11-13.