GeoMMBench - Applications: leaderboard
Metric: Micro-averaged accuracy (%) on the remote sensing application questions of the GeoMMBench test set (expert-written multimodal geoscience and remote sensing multiple-choice and numeric questions, zero-shot, LMMs-Eval, rule-based answer extraction, invalid outputs counted wrong); higher is better. Source: arxiv.org. Saturation forecast: Around July 2028. 36 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 72 |
| 2 | Gemini 2.0 Flash | 64.4 |
| 3 | Qwen 2.5 VL 7B | 64.4 |
| 4 | O1 | 64.4 |
| 5 | Gemini 1.5 Pro | 62.7 |
| 6 | Qwen 2 VL 7B | 61.9 |
| 7 | GPT-5 | 59.3 |
| 8 | Phi-4 Multimodal Instruct | 58.5 |
| 9 | Qwen 2 VL 2B | 47.5 |
Interactive version: theaggregate.ai/benchmark?slug=geommbench-applications · How It Works · Data refreshed daily, snapshot 2026-10-07.