PlantMarkerBench — leaderboard

Metric: Valid F1 (self-reported). Source: benchmarklist.com. 17 models tracked.

Top models

#ModelScore
1Qwen 2.5 32B Instruct75.4
2Mistral 7B Instruct69
3GPT-OSS-20B67.3
4Qwen 2.5 7B Instruct67.1
5Qwen 2.5 1.5B Instruct66.7
6Llama 3.2 1B66.7
7Llama 3.1 8B63.4
8GPT-5.4 Mini63.3
9GPT-5.454.9
10Qwen 2.5 14B Instruct51.4
11Llama 3.2 3B44.9
12Qwen 2.5 3B Instruct24.7

Interactive version: theaggregate.ai/benchmark?slug=plantmarkerbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.