Anthropic has launched Claude Fable 5, the first publicly available version of its advanced Mythos artificial intelligence model. The release marks the first time the company has offered Mythos-derived technology to the general public.
Fable 5 is designed for software engineering, knowledge work, and vision-related tasks. However, Anthropic has imposed restrictions in several high-risk areas. Requests involving cybersecurity, biology, chemistry, and distillation are blocked and redirected to the company’s Claude Opus 4.8 model.
Mythos was first introduced in April as a limited preview available only to selected partners because of cybersecurity concerns. Access was later expanded to hundreds of organizations in 15 countries, primarily operators of critical infrastructure. Anthropic has also announced a newer model, Mythos 5, which is being deployed to approved organizations.
Fable 5 is available through Anthropic’s API and consumption-based Enterprise plans. Until June 22, the model is included at no additional cost for Pro, Max, Team, and seat-based Enterprise subscribers. Beginning June 23, access will require usage credits, although Anthropic says it plans to restore subscription access in the future.
The company said it conducted extensive security testing before release, including more than 1,000 hours of external bug bounty testing and additional red-team evaluations. Anthropic reported that no universal jailbreak methods were discovered.
Anthropic will require 30-day retention of all traffic generated by Fable 5 and Mythos 5 users, including organizations that previously operated under zero-retention agreements. The company said retained data will not be used for model training but for detecting attacks and improving security systems.
Pricing for Fable 5 and Mythos 5 is set at $10 per million input tokens and $50 per million output tokens, double the cost of Opus 4.8. Anthropic said early data shows at least 95% of Fable sessions run entirely on the model without requiring a fallback response.
