MARINER - VQA Naval Equipment Grounding: leaderboard

Maritime visual question answering on MARINER (2026), with multiple-choice questions built from verified image metadata and audited by maritime annotators. Accuracy in percent on naval equipment grounding questions; higher is better.

Metric: Accuracy (%) on MARINER maritime visual question answering (naval equipment grounding questions; multiple-choice questions generated from verified image metadata and audited by maritime annotators); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 16 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Pro58.96#276
2GPT-4.151.89#505
3GPT-4o47.64#751
4Gemini 2.5 Flash45.28#463
5Qwen 2.5 VL 72B Instruct41.98#841
6InternVL3-78B36.79#759
7LLaVA-OneVision-72B35.85#1771
8Qwen 2.5 VL 7B Instruct35.38#1902
9InternVL3-38B35.38#939
10Qwen 2.5 VL 3B Instruct33.02#2713
11Qwen 2.5 VL 32B Instruct28.77#1062
12InternVL3-8B27.83#1782
13LLaVA-v1.5-7B25.94#7630
14MiniCPM-V-2.623.11#3267
15LLaVA-OneVision-7B23.11#2981

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

Interactive version: theaggregate.ai/benchmark?slug=mariner-vqa-naval-equipment-grounding · How It Works · Data refreshed daily, snapshot 2026-11-21.