Media report · AI-ASSISTED BRIEF

OpenAI Releases MedChemBench to Evaluate AI Models on Medicinal Chemistry

Report publication · YYSI publication

According to IT Home, OpenAI released MedChemBench, a 99-question benchmark drawn from 117 papers and patents covering 761 molecules, to test AI models on assembling molecular structures, interpreting evidence, ranking compounds and predicting properties. OpenAI reports GPT-6.2 Astra scored 84.5% at the xhigh reasoning setting, versus 58.2% for GPT-5.6 Sol, with weaker results on compound selection and numeric prediction. OpenAI cautions the results do not establish drug-discovery capability.

Source & context

www.ithome.com · Read the original source ↗

This AI-assisted brief attributes information to the linked media report. It is not an official announcement or independent YYSI reporting. Consult the original report and its evidence.

Report a correction · Editorial standards