Firehose

GPT-Rosalind-6, a frontier reasoning model for biology, drug discovery, and translational medicine, has been evaluated for its biological capability and safety. The model's performance is comparable to GPT-6 Astra, but it exceeds the indicative High capability thresholds on three of the four biological and chemical evaluations. However, none of the four Critical capability evaluations met their indicative thresholds, and the model is treated as High capability in the Biological and Chemical domain, below the threshold for Critical. The model's deployment safeguards include trusted access controls, model-level boundaries, monitoring and enforcement mechanisms, and additional contractual requirements to enhance security. AI summary