AgroVG - Set-F1 at IoU 0.75: leaderboard
AgroVG (2026) agricultural visual grounding scored at a strict box overlap: models box every crop, weed, fruit or pest instance a referring expression names (or none), and the matched box set counts only at IoU 0.75 or more. Query-averaged Set-F1 in percent; higher is better.
Metric: Macro Set-F1 at IoU 0.75 (%, times 100): query-level F1 of the matched box set at the stricter IoU threshold, averaged over all queries, 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: Not forecast. 13 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | InternVL3.5-38B | 25.69 | #1026 |
| 2 | Qwen 3 VL 32B Instruct | 23.93 | #607 |
| 3 | InternVL3.5-8B | 23.87 | #1618 |
| 4 | OWLv2-Large | 23.61 | |
| 5 | Qwen 3 VL 8B Instruct | 21.86 | #1043 |
| 6 | DeepSeek-VL2-Small | 21.79 | #2523 |
| 7 | DeepSeek-VL2-Tiny | 18.17 | #6609 |
| 8 | Grounding DINO 1.5 | 16.59 | |
| 9 | GPT-5.4 | 14.21 | #138 |
| 10 | Florence-2-Large | 13.26 | |
| 11 | GPT-4o (2024-08-06) | 12.34 | #750 |
| 12 | Claude Sonnet 4.6 | 9.97 | #179 |
| 13 | Gemini 2.5 Pro | 7.93 | #280 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=agrovg-set-f1-at-iou-0-75 · How It Works · Data refreshed daily, snapshot 2026-11-22.