GeoMMBench - GIS: leaderboard
Metric: Micro-averaged accuracy (%) on the geographic information system discipline 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 May 2027. 36 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.0 Flash | 86.4 |
| 2 | Gemini 1.5 Pro | 86.4 |
| 3 | GPT-4o | 77.3 |
| 4 | Qwen 2.5 VL 7B | 72.7 |
| 5 | Qwen 2 VL 7B | 70.5 |
| 6 | O1 | 65.9 |
| 7 | Phi-4 Multimodal Instruct | 61.4 |
| 8 | Qwen 2 VL 2B | 59.1 |
| 9 | GPT-5 | 54.5 |
Interactive version: theaggregate.ai/benchmark?slug=geommbench-gis · How It Works · Data refreshed daily, snapshot 2026-10-07.