Anthropic has launched Claude Fable 5, its most capable publicly available AI model, excelling in complex tasks and benchmarks. Alongside it, Mythos 5, a restricted version with fewer safeguards, is available for trusted cybersecurity and life sciences users, aiming to accelerate research and discovery.

Anthropic launches Mythos with Fable 5. Six capabilities for agents: 80.3% code reasoning, extended thinking, autonomy. Vision, science acceleration, enterprise pricing

Anthropic launched Claude Fable 5 on Tuesday, a Mythos-class model with safeguards blocking cybersecurity and biology queries, priced at $10 per million input tokens.