PlantXpert - Quantitative Challenges: leaderboard

PlantXpert (2026) is an evidence-grounded multimodal benchmark for soybean and cotton phenotyping: four-option questions built on figures, captions and text from plant science publications, evaluated on the expert-annotated Gold split with chain-of-thought prompting. This board is accuracy in percent on the questions whose main challenge is quantitative reasoning; higher is better.

Metric: Accuracy (%) on the quantitative reasoning questions on the expert-annotated PlantXpert Gold split (369 four-option questions on soybean and cotton figures from plant science publications), base checkpoints with chain-of-thought prompting, temperature 0, mean of three runs; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen 3 VL 30B A3B55.26#784
2Qwen 3 VL 8B47.37#917
3Qwen 3 VL 4B47.37#1151
4Gemma 3 4B44.74#5339
5Gemma 3 27B39.47#1663
6Gemma 3 12B39.47#2059
7Qwen 3 VL 2B39.47#2134
8InternVL3.5-2B36.84#3284
9Granite-Vision-3.3-2B34.21#6934
10LLaVA-Next-LLaMA3-8B26.32#5654
11LLaVA-NeXT-34B26.32#3443

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

Interactive version: theaggregate.ai/benchmark?slug=plantxpert-quantitative-challenges · How It Works · Data refreshed daily, snapshot 2026-11-21.