Skip to content
Access guide / Anthropic model docs checked October 1, 2026 · plan rules re-checked September 28, 2026 · Kiro model table checked September 29, 2026

Claude Fable 5.1: what changed, what it costs, where to use it

Anthropic released Claude Fable 5.1 on September 1, 2026. What changed from Fable 5, the API price, which Claude plans include it, and where else you can run it.

The short answer

Fable 5.1 is Anthropic’s most capable general-release model, out since September 1, 2026. Per token it costs the same as Fable 5—$10 input and $50 output per million—except cache reads, which drop to $0.25. On Claude plans, Max and premium seats include it up to 50% of weekly limits; Pro uses usage credits.

Check the official documentation (opens in a new tab)

What changed from Fable 5

Anthropic says the gains are largest at higher effort and concentrate in long agentic coding sessions, work on documents, spreadsheets and slides, multistep research, reading dense charts and PDFs, reasoning across the full 1M-token context, and computer use. Because cache reads cost a quarter of Fable 5’s rate, Anthropic estimates about 25% lower cost on typical token-billed work and up to about 45% on highly agentic work. These are the vendor’s own figures.

Official source (opens in a new tab)

Price and specs

API model ID claude-fable-5-1. Per million tokens: $10 input, $50 output, $12.50 for 5-minute cache writes, $20 for 1-hour cache writes, $0.25 for cache reads; batch processing is $5 input and $25 output. It has a 1M-token context window and up to 128k output tokens, and adaptive thinking is always on.

Official source (opens in a new tab)

Which Claude plans include it

Max plans and premium Team/Enterprise seats include Fable 5 and 5.1 up to 50% of weekly usage limits; Pro and Team Standard seats use them with usage credits. Claude Free does not include Fable 5.1, and it was never part of a promotion.

Official source (opens in a new tab)

API, cloud platforms and Kiro

On the Claude API it is available to all customers. It is also on Amazon Bedrock (anthropic.claude-fable-5-1), Claude Platform on AWS, Google Cloud and Microsoft Foundry. Kiro lists Fable 5.1 only as a limited Enterprise preview, enabled by administrators, at a 6× credit multiplier—not on Kiro Free.

Official source (opens in a new tab)

Fable 5.1 or Opus 5.5?

Anthropic’s own guidance is to start most workloads on Opus and move to Fable 5.1 for demanding reasoning and long-horizon agentic work, or when your tests on Opus at higher effort still fall short. Opus 5.5 arrived on September 22 and is included in Pro. Our read: try Opus 5.5 first on your real task; pay for Fable 5.1 where it clearly does better.

Official source (opens in a new tab)

For developers: three breaking changes

Forced tool use (tool_choice “any” or a named tool) returns an error. Earlier models cannot read Fable 5.1’s thinking blocks. Editing earlier turns, the system prompt or the tools list invalidates later thinking blocks, so keep the conversation append-only. Read Anthropic’s migration guide before switching a production integration.

Official source (opens in a new tab)

Questions readers ask

Is Fable 5.1 free?

No. It is not in Claude Free. Max and premium seats include it up to 50% of weekly limits; Pro and Team Standard pay with usage credits; the API bills per token.

Official source (opens in a new tab)
What is the difference between Fable 5.1 and Mythos 5.1?

Anthropic says they are the same model with different safeguards. Mythos 5.1 is limited to approved organizations in its trusted-access programs.

Official source (opens in a new tab)
Is there a Fable 5.2?

Not officially. Leakers say a new Fable is in testing. We track every claim and its source on our Fable 5.2 page.

Official source (opens in a new tab)
Independent guidance, not model access.

Registration here does not create a Claude or Kiro account. Availability and prices can change; provider rules and your account determine what you can use.

Explore the free Jev collection →
Today: OpenAI dots