OpenAI published BioBasicsBench on Nov. 6, a 100-task test built with enzyme-design company Ligo of whether agents can read raw wet-lab evidence, catch errors and stop when the data do not support a conclusion. At the xhigh setting, OpenAI said GPT-6 Astra scored 66.7 percent of available credit, against 43.8 percent for GPT-5.6 Sol. The published comparison covers only OpenAI models, and OpenAI says higher scores do not yet mean reliable lab judgments.
/ Sources
/ About this story
Compiled by Venture Atlas from the sources above, using automated AI-assisted research. This is a summary of reporting published elsewhere, not original reporting - follow the source links for the full account. See our editorial standards.
Something wrong here? Email apostructura@gmail.com and we will fix it.
/ Related
- CNBC and Bloomberg differ on OpenAI's reported $11 billion closeSaturday, November 7, 2026
- OpenAI introduces GPT-6 Cyber for approved Daybreak Red usersThursday, November 5, 2026
- OpenAI sets power rules for campuses it ownsThursday, November 5, 2026
- OpenAI says a cyber model exploited 948 Linux privilege bugsThursday, November 5, 2026