PlantMicro - Localization mAP@0.5: leaderboard
Direct localization in PlantMicro (2026), a benchmark of vision-language models on microscopic plant images: mean average precision in percent of predicted bounding boxes at an IoU of 0.5, zero-shot. Higher is better.
Metric: Mean average precision at IoU 0.5 (%) of directly predicted bounding-box coordinates on microscopic plant images; outputs that violate the format count as failures; zero-shot. Source: arxiv.org. Saturation forecast: Around February 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 56.66 | #167 |
| 2 | Gemini 2.5 Pro | 27.6 | #266 |
| 3 | Gemini 2.5 Flash | 27.01 | #453 |
| 4 | GPT-5 Mini | 23.64 | #348 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=plantmicro-localization-map-0-5 · How It Works · Data refreshed daily, snapshot 2026-11-13.