AgroVG - Single-Target Accuracy: leaderboard

AgroVG (2026) agricultural visual grounding on queries with exactly one target: correct only when the model returns a single box overlapping the target at IoU 0.5 or more. Accuracy in percent; higher is better.

Metric: Strict single-target accuracy (%, times 100): share of single-target queries answered with exactly one box at IoU 0.5 or more, zero-shot agricultural visual grounding (AgroVG-T1): from an image and a referring expression the model returns the boxes of every matching target, an empty list when none is present, on the AgroVG test split (multi-source crop, weed, fruit and pest imagery); predicted and true boxes matched by maximum-cardinality bipartite matching; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 13 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1InternVL3.5-38B35.27#1026
2InternVL3.5-8B34.3#1618
3Qwen 3 VL 32B Instruct31.88#607
4DeepSeek-VL2-Small29.95#2523
5Florence-2-Large25.76
6Qwen 3 VL 8B Instruct24.15#1043
7GPT-5.422.87#138
8Claude Sonnet 4.616.91#179
9Grounding DINO 1.514.17
10Gemini 2.5 Pro14.01#280
11DeepSeek-VL2-Tiny10.31#6609
12OWLv2-Large8.05
13GPT-4o (2024-08-06)4.51#750

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=agrovg-single-target-accuracy · How It Works · Data refreshed daily, snapshot 2026-11-22.