GeoMMBench - Remote Sensing: leaderboard

The remote sensing discipline questions of GeoMMBench (2026), an expert-written multimodal geoscience and remote sensing question benchmark answered zero-shot (test set). Micro-averaged accuracy in percent; higher is better.

Metric: Micro-averaged accuracy (%) on the remote sensing 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 June 2027. 35 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 1.5 Pro68.9#1063
2Gemini 2.0 Flash67.5#723
3GPT-4o65.6#751
4Qwen 3 VL 30B64.5#561
5Qwen 3 VL 8B64#917
6O163.5#403
7Qwen 3 VL 4B61.2#1151
8Qwen 2.5 VL 7B60.6#1862
9GPT-4V59.3#1470
10Qwen 2 VL 7B59.1#2275
11Qwen 2.5 VL 3B56.8#2639
12LLaVA-OneVision-7B55.7#2981
13InternVL2.5-8B55.2#2370
14Phi-4 Multimodal Instruct54#4154
15LLaVA-NeXT-34B52.8#3443

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=geommbench-remote-sensing · How It Works · Data refreshed daily, snapshot 2026-11-21.