CVSBench - CVUSA Satellite-to-Ground: leaderboard

Metric: Accuracy (%) on CVUSA-subset satellite-to-ground questions: from a satellite image, infer spatial attributes of the unseen street view; 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 2030. 14 models tracked.

Top models

#ModelScore
1Qwen 3 VL 4B Instruct62.4
2GPT-4o62.1
3Qwen 3 VL 8B Instruct61.9
4InternVL3.5-8B61.9
5GPT-5 Chat59.5
6Gemma 3 4B (IT)52.88
7DeepSeek V3.247.03

Interactive version: theaggregate.ai/benchmark?slug=cvsbench-cvusa-satellite-to-ground · How It Works · Data refreshed daily, snapshot 2026-09-29.