MARINER - Vessel Detection (Civilian): leaderboard

Vessel detection by multimodal language models on MARINER (2026), a maritime vision benchmark of shore, shipboard, drone and open-source images. COCO-style average precision in percent over civilian vessel categories; higher is better.

Metric: Average precision of MARINER vessel detection by multimodal LLMs (COCO-style average precision over IoU 0.5 to 0.95, percent; civilian vessel categories); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 15 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Pro48.47#276
2InternVL3-38B33.07#939
3InternVL3-78B19.32#759
4MiniCPM-V-2.612.87#3267
5InternVL3-8B11.87#1782
6Gemini 2.5 Flash10.14#463
7GPT-4.19.14#505
8LLaVA-OneVision-72B6.08#1771
9Qwen 2.5 VL 32B Instruct6#1062
10Qwen 2.5 VL 72B Instruct5.64#841
11Qwen 2.5 VL 7B Instruct5.4#1902
12LLaVA-OneVision-7B4.43#2981
13GPT-4o3.12#751
14InternVL2-8B1.81#3227
15Qwen 2.5 VL 3B Instruct0.94#2713

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

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