Anthropic says it is releasing Claude Fable 5.2 and Claude Mythos 5.2, two versions of the same underlying model that differ in who can access them, and it has cut Fable's API price at the same time. CNBC reports that Fable 5.2 is generally available from Monday, November 9. Our earlier report on a Fable 5.5 codename covers a later model name, not this release.

Who can use each model

Fable 5.2 is available on all paid Claude plans, on the Claude Platform and on all major cloud platforms, according to Anthropic's launch thread. Anthropic says Fable carries cyber and biology safeguards, which lets it offer Fable to all users.

Mythos 5.2 is open to organizations that apply through Anthropic's Life Sciences and Cyber Verification Programs. Enterprise customers can use it through Claude Security, according to the launch post. TechCrunch reports that Fable 5.2 had already been available to select parties before release.

Price and speed

  • Fable 5.2: $8 per million input tokens and $40 per million output tokens, down from $10 and $50 for Fable 5.1, per Anthropic's Fable page.

  • Fast mode: up to 2.5 times faster token generation at $16 input and $80 output per million tokens, on the Claude Platform and cloud partners.

  • Speed versus 5.1: Anthropic says both 5.2 models generate outputs up to 30% faster than Fable 5.1 and Mythos 5.1.

At those rates, a job with 1M input and 200k output tokens costs $16 on standard Fable 5.2 and $32 in fast mode, by our arithmetic.

Anthropic recommends starting with Opus 5.5 for most coding and knowledge tasks. It suggests moving up to Fable or Mythos for the most ambitious open-ended technical work, or down to Sonnet 5.5 or Haiku 5.5.

What Anthropic reports on performance

In Anthropic's own benchmark table, Fable 5.2 scored 49.1% on FrontierCode 2.0 at Xhigh effort, against 45.5% for Opus 5.5 and 32.4% for Fable 5.1. Anthropic says Mythos 5.2 found 10.2% more vulnerabilities than Opus 5.5 on its internal CVE Rediscovery evaluation, and 14.4% more than Mythos 5.1.

Anthropic also says Fable 5.2 solved two open problems in theoretical physics about black holes, with preprints linked from its post. It reports that a team of 1,000 Fable 5.2 agents formalized Fermat's Last Theorem in 30 hours, against 11 days for the first formalization. These are Anthropic's own figures.

Safety changes

Anthropic says Fable 5.2 gives about 55% fewer safety-classifier blocks than Fable 5.1 in early testing across claude.ai, Claude Code and Cowork, with about 60% fewer in cybersecurity and 40% fewer in biology.

In a sandbox-escape test, Mythos 5.2 used planted credentials or files outside its sandbox in 3.1% of scenarios, against 39% for Mythos 5 and 21% for Mythos 5.1, according to Anthropic. The company also says both models voice suspicion that they are being evaluated more often than its recent models, which makes real-world behavior harder to assess.

CNBC reports the launch comes as Anthropic readies a potentially historic IPO. It also calls this the first frontier release since CEO Dario Amodei called for a slowdown in AI development.

Our read: for most API users, the Fable price cut and the optional fast mode are the changes that matter. Mythos is limited to vetted organizations, so most readers will only work with Fable.

Sources