Anthropic introduced Claude Fable 5.1 and Claude Mythos 5.1, positioning the release as an upgrade for coding, knowledge work and research-heavy tasks. According to Anthropic’s announcement, the two names refer to the same underlying model, separated by different safeguard regimes: Fable 5.1 is generally available, while Mythos 5.1 is limited to the company’s trusted access programs. The access split is the main product distinction. Anthropic says Mythos 5.1 carries safeguards specifically designed for cybersecurity and life sciences use cases, while Fable 5.1 is the broadly available version for customers using Claude through the company’s products and token-billed channels. Anthropic also framed the release as a response to customer feedback on price, data retention and safety filters. The company says Fable 5.1 will cost an estimated 25% less than Fable 5 for typical workloads wherever usage is billed by token. For highly agentic work, Anthropic says savings will often be larger, reaching up to about 45%. The pricing change comes from lower pricing on cache reads, according to Anthropic. Cache reads apply when the model uses inputs that have already been processed and stored, so the claimed savings should matter most for workflows that repeatedly reference large contexts or run long agentic loops. Anthropic did not provide independent customer-wide cost data in the supplied announcement, so the figures should be read as company estimates. On data handling, Anthropic announced Enterprise Frontier Safeguards, or EFS. The company says EFS is designed to give customers privacy equivalent to zero data retention while preserving safeguards against adversarial use. The mechanism Anthropic describes is customer-controlled cloud infrastructure: data is stored in infrastructure controlled by the customer rather than by Anthropic. EFS is not fully available today. Anthropic says it will roll out to enterprise customers in phases starting later this fall. Until then, eligible customers will be able to use Fable 5.1 with zero data retention, according to the company. The safeguard changes also affect what Anthropic permits the model to do. In cybersecurity, Anthropic says its newest safeguards block 60% fewer false positives than before. The company says Fable 5.1 can now be used to discover software vulnerabilities, but not to develop exploits for them. For biology, Anthropic says it has built an access program in partnership with the US government to enable use of Claude Mythos 5.1’s advanced biology capabilities. The company says it expects to open enrollment for scientists soon, but the supplied materials do not give an exact date or eligibility criteria. Anthropic claims Fable 5.1 improves on Fable 5 across coding, knowledge-work and long-running problem-solving benchmarks, and says lower-effort settings can match or beat Fable 5 at lower cost. It also says Fable 5.1 defaults to High effort in Claude Code and Medium effort in Claude Cowork and on Claude.ai. The announcement includes an early-access anecdote from investment firm Millennium, which Anthropic says used Fable 5.1 to identify the cause of a rare internal-systems crash that had remained unexplained for years. Who benefits: Teams running long-context or agentic workflows may benefit if Anthropic’s estimated cache-read savings hold in production. Enterprise customers with strict data-retention requirements also get a clearer path through EFS and interim zero-data-retention access for eligible users. Who's exposed: Customers that need Mythos 5.1’s cybersecurity or life sciences capabilities remain dependent on Anthropic’s trusted access programs. Enterprises waiting for EFS also face a phased rollout rather than immediate general availability.