← All articles

Anthropic releases Claude Fable 5 with tiered access model

Anthropic launched Claude Fable 5, a publicly available Mythos-class AI model priced at $10 per million input tokens and $50 per million output tokens. The model is free for all subscribers through June 22. A restricted version, Claude Mythos 5, went to cybersecurity partners, while Fable 5 routes sensitive queries to Opus 4.8. Early users including Stripe and IMC have reported significant performance gains.

AI-assisted reporting

The essentials

Anthropic released Claude Fable 5 with tiered access, performance safeguards, and demonstrated capability in complex tasks.

Anthropic has released Claude Fable 5, a Mythos-class AI model available to the public through multiple access tiers. The model is accessible to all subscribers at no additional cost through June 22, with standard API pricing of $10 per million input tokens and $50 per million output tokens thereafter.

The company has implemented a dual-system approach for sensitive queries, routing them to the older Opus 4.8 model while standard requests run on Fable 5. Simultaneously, Anthropic released Claude Mythos 5 to trusted cybersecurity partners without safety restrictions, enabling specialized security research.

Early deployments demonstrate significant capabilities. Stripe completed a 50-million-line Ruby codebase migration in a single day using Fable 5, while IMC trading analysis evaluations show the model excels in financial reasoning tasks. These implementations highlight the model's performance across infrastructure and domain-specific applications.

Broaden the view

Other ways to read this story.

Alternative framings generated by SigNews. These are not independent sources or attributed quotations.

Capabilities

Fable 5 demonstrates advanced performance across diverse domains

The model shows strong performance in specialized domains. Stripe's 50-million-line codebase migration in one day and IMC's success with financial reasoning tasks indicate Fable 5's capability across infrastructure and analytical work. These deployments suggest the model handles both large-scale engineering and domain-specific reasoning effectively.

Access Strategy

Anthropic balances open availability with controlled deployment

The release employs a multi-tier approach: public availability at no cost through June 22, standard API pricing afterward, and unrestricted access for cybersecurity partners. Sensitive queries route to Opus 4.8, providing a safeguard mechanism. This structure enables broad experimentation while maintaining controls for high-risk use cases.