FlagThis

Abstract

This system card describes Claude Mythos 5 and Claude Fable 5, two configurations of a new large language model from Anthropic. Fable 5 is designed for general use and includes safeguards that block tasks in high-risk domains such as biology and cybersecurity. Mythos 5 is a high-capability version with relevant safeguards lifted, available only to a small number of trusted partners. The document details pre-deployment evaluations concerning the Responsible Scaling Policy (RSP), cyber capabilities, agentic safety, and alignment risks.