GeoMMBench - Applications: leaderboard

The remote sensing application 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 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 February 2027. 35 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-4o72#751
2Qwen 3 VL 8B71.2#917
3GPT-4V69.5#1470
4Qwen 3 VL 30B67.8#561
5Gemini 2.0 Flash64.4#723
6Qwen 2.5 VL 7B64.4#1862
7O164.4#403
8InternVL2.5-8B64.4#2370
9Qwen 3 VL 4B64.4#1151
10Gemini 1.5 Pro62.7#1063
11Qwen 2 VL 7B61.9#2275
12GPT-559.3#178
13Phi-4 Multimodal Instruct58.5#4154
14Qwen 2.5 VL 3B57.6#2639
15NVILA-15B57.6#1929

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

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