Measuring how AI agents handle everyday wet-lab science
OpenAI introduces BioBasicsBench, a 100-task benchmark measuring how well AI agents interpret raw experimental data and handle everyday wet-lab science.
Adjust this page to suit you. Your choices are saved in this browser.
Cookies. Chipokia uses one cookie, and only if you allow it: a random ID so your up and down votes stay yours. No analytics, no advertising, no tracking of any kind. Privacy & Cookie Policy.