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
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 48.47 | #276 |
| 2 | InternVL3-38B | 33.07 | #939 |
| 3 | InternVL3-78B | 19.32 | #759 |
| 4 | MiniCPM-V-2.6 | 12.87 | #3267 |
| 5 | InternVL3-8B | 11.87 | #1782 |
| 6 | Gemini 2.5 Flash | 10.14 | #463 |
| 7 | GPT-4.1 | 9.14 | #505 |
| 8 | LLaVA-OneVision-72B | 6.08 | #1771 |
| 9 | Qwen 2.5 VL 32B Instruct | 6 | #1062 |
| 10 | Qwen 2.5 VL 72B Instruct | 5.64 | #841 |
| 11 | Qwen 2.5 VL 7B Instruct | 5.4 | #1902 |
| 12 | LLaVA-OneVision-7B | 4.43 | #2981 |
| 13 | GPT-4o | 3.12 | #751 |
| 14 | InternVL2-8B | 1.81 | #3227 |
| 15 | Qwen 2.5 VL 3B Instruct | 0.94 | #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.