STEMBench120 v1.0.0 (Tools) - Earth science: leaderboard
Metric: Mean fractional scientific progress (0–4; 20 tasks/domain; zero-credit failures included). Source: www.ulam.ai. 1 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-6.1 Sol | 2.68 |
Interactive version: theaggregate.ai/benchmark?slug=stembench120-v1-0-0-tools-earth-science · How It Works · Data refreshed daily, snapshot 2026-10-10.