WildFireVQA - Presence and Detection: leaderboard
Metric: Accuracy (%) on the 8 presence and detection question types of the WildFireVQA wildfire-monitoring multiple-choice questions (34 question types, 6,097 aerial frames each, from FLAME drone imagery with sensor-derived labels), RGB image input without retrieved thermal statistics; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 VL 8B Instruct | 76.39 |
| 2 | InternVL2-8B | 72.79 |
Interactive version: theaggregate.ai/benchmark?slug=wildfirevqa-presence-and-detection · How It Works · Data refreshed daily, snapshot 2026-10-07.