Anthropic released Claude Fable 5 and Claude Mythos 5 on June 9, marking the public debut of its most powerful model family. Fable 5 is the generally available version of the Mythos architecture, delivering what Anthropic calls exceptional performance in software engineering, knowledge work, vision, and scientific research.
Mythos 5, the full-capability version available through the API, demonstrated the ability to autonomously discover and chain zero-day exploits across major operating systems and browsers. To address safety concerns, Anthropic built in hard limits — in high-risk areas like cybersecurity, biology, and chemistry, the model blocks responses and falls back to Claude Opus 4.8. These safeguards trigger in less than 5% of sessions on average.
Both models are priced at $10 per million input tokens and $50 per million output tokens, which Anthropic says is less than half the price of the earlier Mythos Preview. They are available on the Claude API, Claude Platform on AWS, Amazon Bedrock, Vertex AI, and Microsoft Foundry.
The launch comes days after Anthropic published a safety report warning that AI capabilities are advancing faster than safety measures, making the simultaneous release of frontier models and safety guardrails a notable move.