GeoMMBench - Spatial: leaderboard
Metric: Micro-averaged accuracy (%) on the spatial relation 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: Not forecast. 36 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2 VL 7B | 62.2 |
| 2 | Gemini 2.0 Flash | 59.5 |
| 3 | Gemini 1.5 Pro | 56.8 |
| 4 | Qwen 2.5 VL 7B | 54.1 |
| 5 | O1 | 54.1 |
| 6 | Qwen 2 VL 2B | 51.4 |
| 7 | GPT-4o | 48.6 |
| 8 | Phi-4 Multimodal Instruct | 40.5 |
| 9 | GPT-5 | 21.6 |
Interactive version: theaggregate.ai/benchmark?slug=geommbench-spatial · How It Works · Data refreshed daily, snapshot 2026-10-07.