Claude Fable 5.1 sits at the top of Anthropic's fifth-generation Claude lineup, marketed as the most capable option for ambitious coding and professional knowledge work. It is positioned to handle complex, long-running tasks that earlier generations struggled to sustain, with Anthropic framing its research capabilities as an early sign of how AI will contribute to scientific progress. The release is paired with Claude Mythos 5.1, the same underlying model shipped under tighter safeguard constraints for vetted cybersecurity and life sciences users, while Fable 5.1 itself is generally available.

In benchmark reporting from Anthropic, Fable 5.1 sets a new internal standard, scoring 52.6% on Terminal-Bench-Science 0.1 (more than double Fable 5) and 55.8% on Terminal-Bench 4.0 versus 42.0% for Fable 5, and leading every row in a seven-evaluation comparison against Fable 5, Opus 5, and GPT-5.6 Sol. Alongside the capability gains, the rollout introduced a 75% reduction in cache-read pricing, a customer-hosted data retention option, and safeguards retuned to fire far less often on benign requests, building on a prior biology-safeguards update for the Fable 5 line. Together these make Fable 5.1 a strong fit for agentic coding, multi-step research, and other extended professional workloads that benefit from sustained reasoning over very long contexts.