Anthropic launches Claude Fable 5.1 with cost reductions and relaxed safeguards
Anthropic released new AI models aimed at addressing customer complaints about pricing, data privacy, and overly restrictive content filters.
What happened
Anthropic announced Claude Fable 5.1 and Mythos 5.1, claiming the former offers stronger performance than Fable 5 while reducing costs by approximately 25 percent on average and up to 45 percent for complex agentic tasks through reduced pricing on cached data. Fable 5.1 features "more precise safeguards" less likely to block basic biology questions than its predecessor, though Mythos 5.1 maintains the same biology restrictions as the previous model. The company also announced Enterprise Frontier Safeguards that offer "complete privacy" by storing data on customer cloud servers rather than Anthropic's own, with rollout beginning in fall. Anthropic is now permitting Fable 5.1 to identify software vulnerabilities, though certain cybersecurity tasks like penetration testing, exploit generation, and binary-based vulnerability scanning remain restricted to Opus models. Fable 5.1 is available on all platforms; Mythos 5.1 is limited to Project Glasswing participants.
Context
Anthropic's announcement directly addresses reported customer friction points around cost, data retention, and content filtering policies. Early access users from major companies report improvements in performance and usability: Box CEO Aaron Levie noted the model detected ambiguities in data testing that earlier versions missed, and Every CEO Dan Shipper characterized it as strong at coding with improved efficiency and natural communication. The expanded authorization for vulnerability identification represents a controlled expansion of the model's permitted applications in cybersecurity work. The data privacy option addresses enterprise concerns about AI service providers retaining sensitive information.