QUANTSLANT_
Guides

Claude Fable 5.2 Pricing & API: Costs, Cache, Batch, Model ID

quantslant@ai:~$ cat news_1

› Updated: 9 November 2026Claude Fable 5.2 costs $8 per million input tokens and $40 per million output tokens through the Claude API, 20% less than Fable 5.1.

quantslant@ai:~$ cat news_2

› That headline number hides several others that matter for a real bill: cache pricing, the Batch API, fast mode, US-only inference and what the Claude subscription plans include.

quantslant@ai:~$ cat news_3

› I collected all of them from Anthropic's pricing documentation as of 9 November 2026 and worked out what typical requests cost.Claude Fable 5.2 API pricingItemPrice per 1M tokensBase input$8Output$405-minute cache write$101-hour cache write$16Cache hits and refreshes$0.20Batch API input / output$4 / $20Fast mode input / output$16 / $80Three details are easy to miss.

quantslant@ai:~$ cat news_4

› First, cache reads on Fable 5.2 cost 0.025x the input price, not the usual 0.1x, so a cached prompt costs $0.20 per million tokens instead of $8.

quantslant@ai:~$ open --full
Read more

Claude Fable 5.2 Pricing & API: Costs, Cache, Batch, Model ID

Open article on quantslant.com