Anthropic has launched Claude Opus 5.5, an AI model designed to improve performance across biology and other scientific tasks while introducing safeguards intended to address potential misuse. The company said the model exceeds Opus 5 in biology tasks and matches or surpasses Claude Mythos 5.1 across multiple areas of biology.
"Claude Opus 5.5 is our first release since we called for pacing the frontier," Anthropic said in a statement.
The model underwent pre-release testing by external evaluators including Frontier Design and METR. Anthropic also reported that Opus 5.5 performed strongly in its automated behavioral audit, which evaluates model alignment.
Biology capabilities were assessed through a long-horizon molecular prediction and design evaluation conducted with Dyno Therapeutics. Anthropic said the model improved on this evaluation, although it did not disclose a score. Expert red-teamers also assessed the model's scientific novelty.
"Opus 5.5 achieved improvements on a long-horizon molecular prediction and design evaluation conducted in collaboration with Dyno Therapeutics, and expert red-teamers rated its scientific novelty as comparable to the best model they had tested," the company said in a statement.
Opus 5.5 uses the same biology safeguards as Fable 5.1. However, pharmaceutical companies, academic laboratories, startups and other qualified organizations can apply through Anthropic's Life Sciences Verification Program for safeguards designed to accommodate a broader range of legitimate biology research. Applicants undergo review of their research credentials, security standards and ethical oversight.
"As our models grow more powerful, stricter safeguards are one way we prevent new capabilities from becoming tools for misuse," Anthropic said.
Anthropic reported that Opus 5.5 costs 40% less to operate on typical workloads at default settings than Opus 5 and generates output more than 30% faster. The model is available through Anthropic's platforms, Amazon Web Services, Google Cloud and Microsoft Azure.
The launch follows Anthropic's partnership with OpenEvidence to provide AI-supported clinical decision tools to physicians across approximately 100 low- and middle-income countries. Anthropic is providing back-end technology while OpenEvidence adapts its platform to regional healthcare conditions and infrastructure. The companies have not specified whether Opus 5.5 will support that initiative.
Click here for the original news story.
