benchgap
Calibration

DAYJOB Healthcare → DAYJOB

DAYJOB is estimated from DAYJOB Healthcare with a Michaelis–Menten curve fitted on 19 models measured on both: y = 0.4763·x / (0.22251 + x), R² = 0.98, cross-validated error 1.5 pp. It is used for 0 estimates.

Fitted on

ModelDAYJOB HealthcareDAYJOB
Claude Opus 4.80.8%2.0%
Claude Opus 59.2%11.7%
Claude Opus 5.525.2%24.5%
Claude Sonnet 50.0%1.7%
Claude Sonnet 5.516.0%21.7%
Gemini 3.1 Pro0.0%0.0%
Gemini 3.5 Flash-Lite0.0%0.0%
Gemini 3.8 Flash1.2%2.6%
Gemini 3 Flash0.0%0.0%
Gemini 4 Argon11.6%15.7%
Gemini 4 Carbon15.6%19.7%
GPT-5.50.0%1.4%
GPT-5.6 Luna1.2%1.1%
GPT-5.6 Sol1.2%4.2%
GPT-5.6 Terra2.0%2.8%
GPT-6.1 Sol10.8%15.5%
GPT-6 Astra11.2%17.4%
GPT-6 Luna1.2%1.2%
GPT-6 Sol2.8%6.8%