Anthropic has announced the release of Claude Fable 5.1 and Claude Mythos 5.1, described as the world's most advanced models for coding and knowledge work. These models are identical in architecture but differ in their safety guardrails: Fable 5.1 is generally available, while Mythos 5.1 is restricted to trusted access programs, particularly for cybersecurity and life sciences applications.
The new models bring significant improvements in performance, especially in long-running problem-solving tasks. According to Anthropic, Fable 5.1 sets a new standard on coding benchmarks, achieving higher performance than its predecessor, Fable 5, while also being more cost-effective when set to lower effort levels. The models also demonstrate enhanced research capabilities, offering an early glimpse into how AI can contribute to scientific progress.
One of the key updates is pricing: Fable 5.1 will cost an estimated 25% less than Fable 5 for typical workloads due to reduced pricing on cache reads. For highly agentic work, savings could reach up to approximately 45%. Additionally, Anthropic introduced Enterprise Frontier Safeguards (EFS), a new system that gives customers complete privacy by storing data in cloud infrastructure controlled entirely by the customer, not Anthropic. EFS will be rolled out to enterprise customers in phases starting later this fall.
Anthropic also improved safeguards to reduce false positives, particularly in cybersecurity, where the new safeguards block 60% fewer false positives than before. The company has established an access program for advanced biology capabilities in partnership with the US government, with enrollment expected to open soon for scientists.
Early-access partners have reported qualitative improvements, including better writing quality, more effective communication, and the ability to solve complex problems that stumped other models. For example, Fable 5.1 identified the root cause of a rare crash that had eluded engineers for years.