UAVQA-Bench - Quantity Awareness: leaderboard
Metric: Accuracy (%; scene and regional counting and number comparison; direct answering over UAVQA-Bench: 1,500 human-annotated questions on images from 13 public UAV datasets, 16 tasks in six capability dimensions, multiple choice plus box grounding counted correct at IoU 0.5). Source: arxiv.org. Saturation forecast: Around December 2026. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 76.29 |
| 2 | Gemini 3 Flash | 71.71 |
| 3 | Qwen 3 VL 30B A3B (Thinking) | 69.43 |
| 4 | Qwen 3 VL 32B Instruct | 67.71 |
| 5 | Qwen 3 VL 32B (Thinking) | 67.14 |
| 6 | Qwen 3.5 9B | 65.14 |
| 7 | Qwen 3 VL 8B (Thinking) | 64.86 |
| 8 | Qwen 3 VL 8B Instruct | 64.57 |
| 9 | Qwen 3 VL 30B A3B Instruct | 58.57 |
| 10 | InternVL3.5-8B | 53.71 |
Interactive version: theaggregate.ai/benchmark?slug=uavqa-bench-quantity-awareness · How It Works · Data refreshed daily, snapshot 2026-09-29.