Gamow ScreenBench: leaderboard

ScreenBench — Agents inspect compound-screening plate records to detect and correct hidden errors across182tasks. Mean normalized graded-answer score,0no better than shortcuts and1all answers right. Unprompted condition;the prompt listing possible errors is a separate experiment. Native harness/graded population and score revision retained. Native cohorts differ: historical rows use named cell lines; Astra, Carbon and Fable5.2 use coded names; Hydrogen is unspecified. Sol helped select95tasks; evaluation uses fresh runs.

Metric: Mean normalized answer score (fraction, 0–1). Source: gamowlabs.com. Saturation forecast: Around April 2027. 7 models tracked. Updated 15 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Fable 5.20.7026#1
2Gemini 4 HydrogenGamow run: native=vanilla_antigravity__gemini-4-hydrogen-eap, access=EAP0.7024#11
3Gemini 4 CarbonHigh0.6336#2
4Gemini 4 Argon0.4923#12
5Claude Opus 5.5High0.4721#5
6GPT-6.1 SolHigh0.1243#10
7GPT-6 Astra0.1169#6

Interactive version: theaggregate.ai/benchmark?slug=gamow-screenbench · How It Works · Data refreshed daily, snapshot 2026-11-16.