Claude Fable 5.2

Anthropic cut the price with this release. Claude Fable 5.2 costs $8 per million input tokens and $40 per million output tokens, against $10 and $50 for Fable 5.1, and the company says it generates output up to 30% faster than its predecessor. It went on sale on 9 November 2026 under the API identifier claude-fable-5-2.

One underlying model, two ways in

Fable and Mythos are the same underlying model, in Anthropic's words. The difference is the cyber and biology safeguards on Fable, which allow it to be offered to all users, while Claude Mythos 5.2 is reserved for approved security and life sciences organisations. Anthropic steers most coding and knowledge work to Claude Opus 5.5 and presents Fable for ambitious, open-ended technical tasks, including consulting for or delegating to smaller models.

Context, speed and where it runs

The context window is 1,000,000 tokens and maximum output 128,000, with text and image input and text output. Adaptive thinking is always on, the default effort is high, and the knowledge cutoff is June 2026. An optional fast mode generates tokens up to 2.5 times faster at $16 input and $80 output. Anthropic offers the model on its own platform and through Amazon Web Services, Google Cloud and Microsoft Azure, and it says retirement will come no sooner than 9 November 2027.

What outsiders have measured

Artificial Analysis gives Fable 5.2 a score of 65 on its Intelligence Index, first of the 74 models in its comparison when read on 10 November 2026, at a cost of $26.54 per index task and an output speed of 98 tokens a second. Anthropic's own claims about migrations, device design and physics results are its own and are not independently tested.

Key facts

Maker: Anthropic Released: 9 November 2026 Model code: claude-fable-5-2 Context window: 1,000,000 tokens Maximum output: 128,000 tokens Input and output: text and image in, text out Price per million tokens: $8 input, $40 output Availability: Claude Platform, Amazon Web Services, Google Cloud, Microsoft Azure