MARINER - Vessel Detection (Military): leaderboard

Metric: Average precision of MARINER vessel detection by multimodal LLMs (COCO-style average precision over IoU 0.5 to 0.95, percent; military vessel categories); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 13 models tracked.

Top models

#ModelScore
1Gemini 2.5 Pro69.04
2Qwen 2.5 VL 72B Instruct37.29
3Qwen 2.5 VL 32B Instruct26.91
4Qwen 2.5 VL 7B Instruct26.1
5Gemini 2.5 Flash22.91
6InternVL3-38B21.77
7InternVL3-78B7.91
8MiniCPM-V-2.65.93
9InternVL3-8B2.37
10InternVL2-8B0.43

Interactive version: theaggregate.ai/benchmark?slug=mariner-vessel-detection-military · How It Works · Data refreshed daily, snapshot 2026-10-07.