Skip to content
Climb Google Maps: Google Business Profile optimization.

Automation

Claude Fable 5.2 Is Out: Cheaper, Faster, and Open to Every Plan. What It Means for People Who Build Websites

Anthropic released Claude Fable 5.2 on Nov. 9 at $8 per million input tokens, with a fast mode and extra usage for paid plans this week. Here is what changed, what it costs, how to switch to it today, and which of Anthropic's numbers matter to marketers and site builders.

A processor chip mounted on a computer circuit board

Key takeaways

  • Anthropic released Claude Fable 5.2 and Claude Mythos 5.2 on Nov. 9, 2026; they are the same underlying model, and Fable 5.2 is the version with stronger safeguards that anyone can use.
  • Claude Fable 5.2 costs $8 per million input tokens and $40 per million output tokens on the API, twice the price of Claude Opus 5.5, with a fast mode at $16 and $80.
  • Anthropic says Fable 5.2 generates outputs up to 30% faster than Fable 5.1, and fast mode speeds up token generation by up to 2.5 times.
  • In Claude Code, Fable is never the default model: run /model fable or claude --model fable, on version 2.1.321 or later.
  • Anthropic's own benchmarks put Fable 5.2 ahead of Opus 5.5 on agentic coding (69.9% vs. 62.3% on FrontierSWE v2), while Gemini 4 Carbon scored higher on its business-workflow test.

Anthropic released Claude Fable 5.2 on Nov. 9, 2026, alongside Claude Mythos 5.2. Anthropic calls them its most capable models. Fable 5.2 is available to everyone today, in the Claude apps, Claude Code, the API, and through Amazon Web Services, Google Cloud and Microsoft Azure.

For anyone who builds or runs websites with AI, three things matter: it is cheaper than the Fable model it replaces, it is faster, and paid Claude plans get extra Fable usage this week. It is also still twice the API price of Claude Opus 5.5, so it is a model to pick for the hard jobs, not for everything.

What is the difference between Fable 5.2 and Mythos 5.2?

Anthropic says they are the same underlying model. Fable 5.2 carries stronger safeguards around cybersecurity and biology, which is why it can be offered to all users. Mythos 5.2 is limited to vetted organizations through Anthropic's Cyber Verification and Life Sciences Verification programs.

When Fable's safeguards block a request, Anthropic says the answer comes from another model instead: Opus 4.8 for cybersecurity and Opus 5 for biology. Anthropic says blocks are down about 55% from Fable 5.1. For website and marketing work, these safeguards should rarely come up.

How much does Claude Fable 5.2 cost?

On the API, from Anthropic's announcement and model documentation:

ModelInput, per million tokensOutput, per million tokens
Claude Fable 5.2$8$40
Claude Fable 5.2, fast mode$16$80
Claude Opus 5.5$4$20
Claude Sonnet 5.5$2$10

Anthropic describes the $8 and $40 as "a lower price" than Fable 5.1. Batch requests are half price, and the documentation lists prompt cache reads on Fable 5.2 at 2.5% of the base input price, which matters if you send the same long instructions or style guide with every request.

On subscriptions, Anthropic says this week eligible paid plans get "one to three Fable resets" depending on how long they have subscribed, and that Fable is included for up to 50% of weekly usage limits on Pro plans and standard Team seats. Claude Code's documentation adds that on some plans and seats Fable usage bills to usage credits instead; the model picker shows "Requires usage credits" when that applies.

What changed from Fable 5.1?

  • Speed. Anthropic says outputs are up to 30% faster than Fable 5.1. The optional fast mode speeds up token generation by up to 2.5 times, at double the price. It is available in Claude Code, the Claude Platform, Claude Models on Google Cloud and Microsoft Foundry.
  • Fewer tokens by default. Medium effort is now the default in the Claude apps and Claude Code, and Anthropic says it uses fewer tokens than Fable 5.1 did.
  • Coding judgment and agents. Anthropic highlights better engineering judgment on vague requests and better coordination of subagents, the helper agents a lead agent hands parts of a job to.
  • Specs. The model documentation lists a 1 million token context window, up to 128,000 output tokens, and a knowledge cutoff of June 2026. The API model ID is claude-fable-5-2.

How do you use Fable 5.2 in Claude Code?

Fable is not the default model on any plan, so you have to choose it. From Claude Code's model documentation:

  • In a session, type /model fable. Typed this way, the choice is saved as your default for new sessions. To switch for one session only, type /model, then press s in the picker.
  • At launch: claude --model fable.
  • It needs Claude Code version 2.1.321 or later; run claude update first.
  • Effort runs from low to max. The documentation suggests medium for day to day work and xhigh for deeper reasoning.

On the API, use the model ID claude-fable-5-2. If you are not sure which model to use, the documentation's own advice is to start with Opus 5.5 and move to Fable 5.2 for "demanding reasoning and long-horizon agentic work," or when Opus 5.5 at a higher effort still falls short.

Which of Anthropic's numbers matter for website work?

These are Anthropic's own results, not tests we ran:

  • Agentic coding: 69.9% on FrontierSWE v2, against 62.3% for Opus 5.5. This is the closest measure to "build and fix a real site."
  • Computer use: 89.3% on OSWorld 2.1 (offline set), against 87.2% for Opus 5.5. That is the skill an agent uses to click through a CMS or an ad platform in a browser.
  • Business workflows: 51.4% on AutomationBench. Here Google's Gemini 4 Carbon scored higher, at 53.9%, so Fable 5.2 is not the leader on every task.

Anthropic also says that over 90% of the changes merged into its core codebase are now written by Claude, "with human guidance." That last part is the lesson for site owners too: the model writes, a person decides.

Our view: who should switch, and for what

Use Fable 5.2 where a mistake is expensive and the job is long: a full site build or redesign, a migration, a technical SEO cleanup across hundreds of templates, or a long agent run that needs to plan, check its own work and keep going. Keep Opus 5.5 or Sonnet 5.5 for everyday edits, single pages, ad copy variations and anything that runs at volume, where Fable's price doubles your bill for little gain.

This site runs on Claude Opus 5.5 today, and Fable 5.2 is what we would test next on the hardest parts of our pipeline. We explain how we build and run websites with Claude, and our step by step guide shows how to build a website with Claude using Fable 5.2 or Opus 5.5.

What we don't know yet

  • Real cost per job. Anthropic's speed and token savings are its own measurements. How much a typical site build costs on Fable 5.2 compared with Opus 5.5 depends on the job, and we haven't measured it.
  • How long the extra plan usage lasts. Anthropic frames the extra Fable usage as "this week." Check the model picker in Claude Code before planning a long build around it.

Sources

Photo of Kiran Dasgupta

About the author

Kiran Dasgupta, Editor

Kiran runs Kihan Marketing, a web design, SEO and ads agency, and manages dozens of websites across real estate, home services and healthcare. He writes about SEO, AI search and paid ads news, and what it means for the people who run websites.