AgroTools - Tool Selection Accuracy: leaderboard

Step-by-step mode on AgroTools (2026), image-grounded agricultural questions (pest and disease diagnosis, counting and measurement, change detection, visualization) with 14 executable agricultural tools: given the gold history, the model predicts the next step, and this is the share of gold tool-call steps where it calls the reference tool. Percent; higher is better.

Metric: Tool accuracy (%): share of gold tool-call steps where the model calls a tool and selects the reference tool, in step-by-step mode on AgroTools (539 image-grounded agricultural queries with human-specified, human-audited reference tool trajectories over 14 executable tools; the model predicts the next step given the gold history), Lagent ReAct schema, temperature 0.1; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 13 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Seed 2.0 Pro95.16#222
2Gemini 2.5 Pro91.91#276
3Qwen 3.5 35B A3B91.71#492
4Qwen 3.5 9B86.66#828
5InternVL3.5-8B85.49#1604
6InternVL3.5-14B85#1504
7GLM-4.6V84.93#710
8Qwen 3 VL 8B Instruct67.31#1029
9GPT-5.452.32#135
10Claude Sonnet 4.650.79#173
11LLaVA-v1.6-Vicuna-13B44.85#5527
12InternVL3-8B20.66#1782
13Janus-Pro-7B15.96#6857

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

Interactive version: theaggregate.ai/benchmark?slug=agrotools-tool-selection-accuracy · How It Works · Data refreshed daily, snapshot 2026-11-21.