Anthropic launches Claude Fable 5.1 (public) and Mythos 5.1 (restricted)
Anthropic released two same‑core models with different safeguard profiles: Fable 5.1 for broad developer access, and Mythos 5.1 for vetted cyberdefense and life‑science use under trusted access programs and enterprise safeguards (EFS).
In this brief: 3 sections 2 min read
Fable 5.1 and Mythos 5.1 are the same underlying model with different safeguard configurations.
Claude Fable 5.1 is generally available today across Anthropic’s platforms and cloud partners (AWS, Google Cloud, Microsoft Azure) and via the Claude API (claude-fable-5-1).
Claude Mythos 5.1 is only available through Anthropic’s trusted access programs for vetted cyberdefense and life‑science organisations; availability is currently limited to a set of US organisations while Anthropic coordinates with US government partners to expand access.
Anthropic described Enterprise Frontier Safeguards (EFS) to let eligible customers store data on their cloud infra and receive zero data retention while enabling customer‑conducted human review by default.
Claude Mythos 5.1 has more permissive safeguards tailored for cybersecurity and life‑sciences work and is being used to power Anthropic’s Claude Security code‑scanning product.
EFS rollout will begin in phases starting this fall; customers can request access via Anthropic’s form.
Developers can access claude-fable-5-1 on the Claude API and through cloud partners listed in the announcement.
Organisations seeking Mythos 5.1 for cyberdefense are directed to register interest through Anthropic’s CVP process; broader international access is being coordinated with government partners.
Anthropic highlighted discounted support for scientists and grants through its AI for Science program and a Claude Team plan for researchers.