PlantXpert - Reasoning 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 multi-step agronomic reasoning; higher is better.

Metric: Accuracy (%) on the multi-step agronomic 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
1Gemma 3 12B70.73#2091
2Qwen 3 VL 30B A3B70.73#791
3Qwen 3 VL 8B69.92#927
4Qwen 3 VL 4B69.92#1163
5Gemma 3 27B65.04#1672
6InternVL3.5-2B62.6#3313
7Granite-Vision-3.3-2B58.54#7001
8LLaVA-NeXT-34B54.47#3483
9LLaVA-Next-LLaMA3-8B50.41#5702
10Qwen 3 VL 2B48.78#2156
11Gemma 3 4B47.97#5347

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

Interactive version: theaggregate.ai/benchmark?slug=plantxpert-reasoning-challenges · How It Works · Data refreshed daily, snapshot 2026-11-22.