Introducing Claude Fable 5.1 and Claude Mythos 5.1

September 1, 2026 | Source: Anthropic News (community RSS)

Tags: Claude Fable 5.1, Claude Mythos 5.1, Anthropic, model release, enterprise AI, Claude Code, agentic AI, EFS

Anthropic releases Claude Fable 5.1 — its most capable coding and knowledge-work model — at 25% lower cost for standard workloads (up to 45% for agentic pipelines), with customer-controlled data storage for regulated enterprises and a 60% drop in cybersecurity false positives.

Details

Claude Fable 5.1 and Mythos 5.1 are the same underlying model with different safeguard tiers. Fable 5.1 is generally available; Mythos 5.1 is restricted to trusted access programs for cybersecurity and life sciences. Both are Anthropic's most capable models for coding and sustained problem-solving. Three enterprise-facing changes anchor the release. Pricing drops roughly 25% for typical workloads via reduced cache-read costs; highly agentic pipelines see savings up to 45%. The new Enterprise Frontier Safeguards (EFS) system stores interaction data in customer-controlled cloud infrastructure (AWS, Google Cloud, or Azure), combining zero data retention privacy with cross-session misuse monitoring — rolling out in phases this fall. In the meantime, eligible customers get zero data retention on Fable 5.1 as a bridge. Cybersecurity safeguards now flag 60% fewer false positives, and Fable 5.1 can assist with vulnerability discovery — though not exploit development. Anthropic also highlights agentic scientific research capabilities, scored on a new Terminal-Bench-Science 0.1 benchmark. A biology capabilities program for Mythos 5.1 is being developed in partnership with the US government. Effort levels from Low to Max let users tune cost vs. performance; Claude Code defaults to High, Claude.ai to Medium.