Skip to main content

Claude Fable 5.2 Launches: Up to 30% Faster, Priced at $8/$40 per Million Tokens

Anthropic's official account @ClaudeDevs announced on November 9, 2026 that Fable 5.2 is available in the Claude Platform and Claude Code. It generates output up to 30% faster than Fable 5.1 and is priced at $8 input / $40 output per million tokens. Launch-week Fable usage offers were announced alongside it.

By
🌐 This article was machine-translated and may contain inaccuracies. Read the Korean original if in doubt.

Anthropic's official developer account @ClaudeDevs announced the following on X on November 9, 2026:

Key pointsFable 5.2 launches, up to 30% faster than 5.1$8 input / $40 output per 1M tokensFast mode: up to 2.5x speed at $16/$80Launch Fable usage resets for eligible plans
Fable 5.2 is now available in the Claude Platform and Claude Code. It’s up to 30% faster than Fable 5.1, and costs $8/$40 per 1M tokens, down from $10/$50.

On the same day Anthropic published an official page introducing Fable 5.2 and Mythos 5.2, platform documentation, and a launch-offer support article. Everything below is limited to what those sources say.

What Was Announced

  • Speed — Fable 5.2 and Mythos 5.2 generate output up to 30% faster than Fable 5.1 and Mythos 5.1. Because the wording is 'up to', the gain will not necessarily be the same on every task.
  • Price — $8 per million input tokens and $40 per million output tokens. According to the post, the previous price was $10/$50.
  • Fast mode — an optional mode that speeds up token generation by up to 2.5 times, priced at $16 input and $80 output per million tokens.
  • Fable and Mythos — they are the same underlying model. Fable has strong cybersecurity and biology safeguards, which lets Anthropic offer it to all users. Organizations doing beneficial work in those fields can apply for Mythos access through the expanded Life Sciences and Cyber Verification Programs.
  • Efficiency — at Medium effort (a setting that controls how much the model thinks), Fable 5.2 handles challenging tasks with fewer tokens than Fable 5.1. In early FrontierCode 2.0 testing, Fable 5.2 at High effort scored about level with Opus 5.5 at Max effort, for roughly half the cost per task.
  • Safety — Anthropic ran extensive prerelease safety testing and says high-stakes risks from misalignment remain low. Its cyber and biology safeguards are as robust as Fable 5.1's but produce fewer unnecessary blocks. Details are in the system card.

The official page has a benchmark table (FrontierCode 2.0, FrontierSWE v2, OSWorld 2.1 and others) presenting Fable 5.2 as a step above Opus 5.5 in almost all evaluated capabilities. For exact figures and methodology, see the official page and the system card.

What It Means for Users

Lower prices and faster output reduce the cost of using Fable for hard, long-running work. Anthropic still recommends that most coding and knowledge tasks start with Opus 5.5, then move up to Fable or Mythos, or down to Sonnet 5.5 or Haiku 5.5, depending on scope and complexity. Fable 5.2 can also serve as an Advisor (Haiku, Sonnet, or Opus models consult it on harder problems) or an Orchestrator (it delegates simpler tasks to smaller models).

How to Use It

Requirements and Where It Is Available

  • Availability — per the post, in the Claude Platform and Claude Code. The platform docs list the Claude API, Amazon Bedrock, Claude Models on Google Cloud (Google managed and Anthropic managed), Microsoft Foundry, and Claude Platform on AWS.
  • Model IDs — claude-fable-5-2 on the Claude API and anthropic.claude-fable-5-2 on Amazon Bedrock.
  • Mythos 5.2 — offers the same capabilities only to organizations verified through Anthropic's programs, such as the Cyber Verification Program. Apply to the program covering your use case or contact your account team.

Key Specifications

  • Context and output — 1M-token context window, 128K max output; text and images in, text out.
  • Thinking — adaptive thinking (the model adjusts its thinking depth as needed) is always on and cannot be turned off. Depth is controlled with the effort parameter; the API default is high, while the default in Claude apps and Claude Code is Medium.
  • Knowledge cutoff — both the reliable knowledge cutoff and the training data cutoff are June 2026.

Pricing Details

  • Base — $8 / MTok input, $40 / MTok output (MTok = million tokens).
  • Caching — 5-minute cache write $10, 1-hour cache write $16, cache read $0.20 / MTok.
  • Batch — the Batch API gives a 50% discount on input and output.
  • Fast mode — offered for Messages API requests at premium pricing. The exact setup steps need to be confirmed in the Fast mode page of the platform docs.

API Cautions

  • Forced tool use (tool_choice types any and tool) returns a 400 error.
  • Omit temperature, top_p, and top_k.
  • 30-day data retention is required, and the model is not available under zero data retention unless expressly authorized by Anthropic.
  • Editing earlier turns invalidates thinking blocks.

Launch Offers (Claude App Subscribers)

  • Pro plans and standard seats on Team plans — all Fable models are included from November 9, 2026 through November 16, 2026 at 11:59 pm PT, up to 50% of weekly usage limits, plus one Fable limit reset that expires November 16, 2026 at 11:59 pm PT.
  • Max and Ultra plans, premium seats on Team plans, and premium seats on legacy seat-based Enterprise plans — one to three Fable resets depending on subscription length (under six months: one; six to 12 months: two; over 12 months: three). They expire March 8, 2027 at 11:59 pm PT.
  • Eligibility — paid billing must have been in place before November 9, 2026 at 10:00 am PT. The Free plan, standard seats on legacy seat-based Enterprise plans, self-serve Enterprise plans, and usage-based Enterprise plans are not eligible.
  • How to use a reset — go to Settings > Usage, and under Limit resets select “Reset for free.” Fable usage is restored right away. You can use it any time before it expires, not only at a limit, and it can't be undone.
  • Notes — your default model doesn't change, so select a Fable model from the model picker. Fable models use limits faster than other Claude models. A reset does not raise your limits and has no cash value or transferability.

Who Uses It for What

The platform docs name agent sessions that run for hours, multistep deep research, and analysis carried through to a finished document, spreadsheet, or deck as typical uses.

What Else to Check

  • The specific command or menu for selecting Fable 5.2 in Claude Code, and plan-specific Claude Code usage limits, were not included in these materials. Please confirm them in official sources.
  • The previous $10/$50 price appears only in the post; the official pages provided do not list it separately.
  • Benchmark conditions (safeguards on or off, effort level) differ by item. Read the footnotes on the official page and the system card when comparing.

Prices, availability, and offer terms may change, so please check Anthropic's official channels for the latest status.

Official resources: anthropic.com/claude-fable-and-mythos-5-2 · platform.claude.com/overview · support.claude.com/17506494-claude-fable-5-2-launch-offers

Was this helpful?

Keep reading