UAVBench - Classification (MOD20): leaderboard
Metric: Image-level action classification accuracy (%), zero-shot with the authors' task instructions (no fine-tuning on UAVIT-1M), on the UAVBench test unit built from the test split of MOD20 (out-of-domain), low-altitude drone imagery; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash | 97.5 | #237 |
| 2 | Qwen 2.5 VL 7B Instruct | 96.17 | #643 |
| 3 | MiniCPM-V-2.6 | 84.67 | #825 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=uavbench-classification-mod20 · How It Works · Data refreshed daily, snapshot 2026-10-11.