Claude Fable 5.1 Is Now Available

Anthropic released Claude Fable 5.1 today. It is now in the AI Data Stream model picker as the top-tier premium Claude option. Fable 5 remains available as the previous generation, so you can compare the two on your own analytics before switching.

Pricing

Input and output prices are unchanged from Fable 5:

• Input: $10 per million tokens
• Output: $50 per million tokens

What did change is the price of a cache read: $0.25 per million tokens, down from $1. That sounds like a footnote, but it is the number that matters most for the way this app works. Every time you send a follow-up question, the whole conversation so far is re-read from cache. In a long analysis session that cached prefix is the bulk of what you pay for, so a quarter of the price on cache reads takes a real bite out of the cost of a long conversation.

For context, Fable 5.1 cache reads are now cheaper than Opus 5’s ($0.50) — on the single most repeated part of a conversation, the most expensive model in the picker is no longer the most expensive to keep talking to.

What’s New

Anthropic reports gains concentrated in long-horizon work: agentic coding across sessions that run for hours, multistep research that follows up on what it finds, and reasoning across the full 1M-token context window. Two of the improvements land squarely on analytics work — vision, specifically reading dense charts, filings, and tables nested inside PDFs, and document and spreadsheet work, taking an analysis from a first question through to a finished document. The gap over Fable 5 is widest at the higher effort levels. See Anthropic’s announcement for the full details.

Anthropic also notes Fable 5.1 writes denser prose and uses less formatting in chat than earlier Claude models, so expect its reports to read a little more like continuous argument and a little less like a bulleted list.

Fable 5 Has Been Kept

Nothing has been retired. Fable 5 stays in the picker labelled as the previous generation, and if it was your default it stays your default — no action needed, no automatic migration. Sonnet 5 remains the default model for new providers.

Effort Levels

Fable 5.1 supports the full effort scale: low, medium, high, xhigh, max, defaulting to high. Anthropic’s guidance is worth repeating here: the improvement over Fable 5 is largest at the top of that scale, and at low effort Fable 5.1 is more inclined to answer from memory instead of reaching for a tool. For questions that need fresh numbers pulled from your connected data, keep it at high or above.

A Note on Data Retention

Like Fable 5, Fable 5.1 requires 30-day data retention on your Anthropic account and is not available to organisations configured for zero data retention. If your Anthropic account is set up that way, Fable 5.1 requests will be rejected — use Opus 5 or Sonnet 5 instead.

The Full Claude Lineup

• Claude Fable 5.1 — Most powerful, premium; for the hardest reasoning
• Claude Fable 5 — Previous-generation Fable
• Claude Opus 5 — Frontier model, for demanding tasks
• Claude Opus 4.8 — Previous-generation Opus
• Claude Sonnet 5 — Best balance of speed and capability (default)
• Claude Sonnet 4.6 — Previous generation
• Claude Haiku 4.5 — Fastest, great for quick queries

Switching Models

1. Go to Settings > AI Providers
2. Click your Anthropic provider
3. Select your preferred model from the dropdown
4. Save

Or select it from the model picker when starting a new conversation.

Questions?

If you have any questions, contact us or check our AI providers documentation.