Claude Fable 5 Lands, Then Pauses: Anthropic's Most Powerful Public Model
On June 9, 2026, Anthropic released Claude Fable 5, which it describes as its most capable model made safe for general use. Within days it was temporarily suspended over a U.S. government export directive, according to reporting from InfoQ. If you saw Fable appear and then vanish from a model picker, that is why.
What Fable 5 actually is
Fable 5 is the public, guardrailed sibling of Claude Mythos, the model Anthropic disclosed on April 7, 2026, and chose not to release. Per Anthropic, Fable 5 and Mythos 5 share the same underlying model and specifications; the difference is that Fable ships with safety classifiers that can decline requests, while Mythos does not. In high-risk domains such as cybersecurity, biology, and chemistry, Fable is reported to block the response and fall back to Claude Opus 4.8.
It carries a 1 million token context window and up to 128,000 output tokens per request. Reported pricing is around 10 dollars per million input tokens and 50 dollars per million output tokens, roughly double Opus 4.8 on both sides.
Does it matter for vibe coding
For most day-to-day building, no, and that is fine. Fable is built for the hardest, longest-horizon reasoning and agentic work, and it costs about twice as much as Opus 4.8. Opus 4.8 is still the sensible default for everyday vibe coding. Reach for a model like Fable only when a task genuinely exceeds what the default can hold, a large multi-step migration or a deep reasoning problem, and the budget justifies it.
The bigger signal is the trajectory: the frontier models are now explicitly tiered by capability and risk, and the most powerful one shipped with a kill switch attached. Worth watching, not worth re-tooling your workflow over.