Anthropic announced the general availability of the Claude Fable 5.1 model. ^[1] Anthropic launched Claude Fable 5.1 and Mythos 5.1, its most capable AI models to date, which share the same base model but differ in safety guardrails. ^[2] Anthropic released a new model named Mythos 5.1 specifically for trusted partners. ^[3] The Mythos 5.1 model includes specific safeguards designed for cybersecurity applications. ^[4] Anthropic reduced the cost of cache reads from $1 to $0.25 per million tokens to achieve the price reductions, while keeping other API prices unchanged at $10 per million input tokens and $50 per million output tokens. ^[5] Anthropic stated that the new models can cut costs by up to 45 percent, with Fable 5.1 costing about 25 percent less than Fable 5 for typical workloads. ^[6] On the Terminal-Bench-Science 0.1 benchmark, Fable 5.1 scored 52.6 percent, more than double the 24.7 percent score of Fable 5. ^[7] There is a reported 5.1-point gap between Claude Fable 5.1 and Mythos 5.1. ^[8] Anthropic claims Fable 5.1 will cost about 25% less than Fable 5 for typical token-based workloads, with savings up to approximately 45% for highly agentic work. ^[9] Anthropic claims Mythos 5.1 developed full working exploits in 98% of trials, up from 88%. ^[10] The new models feature an increased token appetite that is expected to shorten user sessions. ^[11]