Claude Fable 5.2 API pricing: $8 input, $40 output per million tokens

What the Claude pricing page shows for Fable 5.2

As of November 9, 2026, Claude's pricing page lists Fable 5.2 first among its latest models and describes it as next generation intelligence for long-running agents. Its listed API rates are $8 per million input tokens (MTok) and $40 per MTok of output. Prompt caching costs $0.20 per MTok for cache reads and $10 per MTok for cache writes. The page says these caching prices are for a 5-minute TTL.

Fable 5.1 is now listed under legacy models at $10 per MTok input, $50 per MTok output, $0.25 per MTok for cache reads and $12.50 per MTok for cache writes. Fable 5.2's listed input, output, cache read and cache write prices are each 20% lower than Fable 5.1's. The page does not give a date for the change or a retirement date for Fable 5.1.

Sources: Pricing

Fast mode now covers Fable 5.2 at double the price

The pricing page says fast mode is available for Opus 5.5 and Fable 5.2, with speeds up to 2.5x faster at 2x standard pricing. At Fable 5.2's listed rates, that works out to $16 per MTok input and $80 per MTok output.

Sources: Pricing

What this means for your bill and plan usage

On the API, Fable 5.2 requests are billed at the per-token rates above, or at double those rates in fast mode. On subscription plans, the same page says Pro gets Fable paid with usage credits. Max and Ultra include Fable for up to half of the weekly limit. On paid plans, usage credits let you keep working at standard API rates after you reach a limit.

The pricing page does not say which Fable version the Pro, Max, Ultra or Team plans use. It also does not say whether Fable usage paid with usage credits is now charged at Fable 5.2 rates.

Sources: Pricing