Claude Fable 5: Anthropic's Most Powerful AI Model Yet

Anthropic has launched Claude Fable 5, its most powerful public AI model, making Mythos-level artificial intelligence available to enterprise customers and paid subscribers. This new model delivers exceptional performance in software engineering and knowledge work. Here's the thing: Fable 5 scored 80.3% on the SWE-Bench Pro software engineering benchmark and was the first AI model to cross the 90% threshold on Hex's analytical benchmark. It also outperformed Claude Opus 4.8 by more than 10% on other tests. What's interesting is that Fable 5 includes built-in safety guardrails. In high-risk areas like cybersecurity, biology, and chemistry, it automatically blocks responses and reverts to Claude Opus 4.8. Early data shows 95% of Fable sessions use the new model's responses without triggering restrictions. The model is designed for ambitious, multi-day tasks. Fable 5 is priced at $10 per million input tokens and $50 per million output tokens. It's available on Anthropic's platform and through major cloud providers. This launch significantly shifts the AI landscape.