Anthropic releases Claude Fable 5.2 and Mythos 5.2 at $8 / $40, 20% below 5.1, and Artificial Analysis ranks Fable 5.2 first at 65

Anthropic released Claude Fable 5.2 and Claude Mythos 5.2 on November 9. Fable 5.2 is available now to everyone on the Claude API as claude-fable-5-2, and on Amazon Web Services, Google Cloud and Microsoft Azure. Mythos 5.2 is the same underlying model without certain cybersecurity and biology safeguards, and only organizations in Anthropic's verification programs can use it. Our radar first saw claude-fable-5-2 at 18:52:10 UTC and sent the alert at 18:53:06, 4 minutes 53 seconds before @claudeai announced it.

THE PRICE AND SPECS

• $8 input / $40 output per 1M tokens, down from $10 / $50 for Fable 5.1 and Mythos 5.1.
• An optional fast mode, which Anthropic says generates tokens up to 2.5 times faster, costs $16 / $80 per 1M. For Mythos it is open only to verification-program members.
• 1M-token context window, 128K max output, adaptive thinking always on and a reliable knowledge cutoff of June 2026, per Anthropic's models overview.
• Default effort is high on the API and medium in Claude apps and Claude Code.

WHAT ANTHROPIC SAYS

These are Anthropic's own figures:

• Output up to 30% faster than Fable 5.1 and Mythos 5.1.
• FrontierCode 2.0: at High effort Fable 5.2 scores 45.3%, about level with Opus 5.5's best of 45.5%, for roughly half the cost per task. At Xhigh it reaches 49.1%, which Anthropic says is the highest of any released model it tested. GPT-6 Astra's best was 42.8%.
• BenchCAD 2.1 (hardware design): 83.5 at Max effort, against 65.3 for Opus 5.5 and 44.4 for Fable 5.1.
• RoboDojo Sim (robot control): 64.9%, against 29.8% for GPT-6 Astra and 27.6% for Fable 5.1.
• A team of about 1,000 Fable 5.2 agents formalized the proof of Fermat's Last Theorem in 30 hours.
• Fable 5.2's safety classifiers block most cybersecurity and research-biology tasks; Mythos 5.2 is the version for that work.
• Anthropic still recommends starting with Opus 5.5 for most coding and knowledge work, and moving up to Fable or Mythos for the most ambitious technical work.

THE INDEPENDENT NUMBER

Artificial Analysis scores Claude Fable 5.2 (Max) 65 on its Intelligence Index and ranks it first of the 74 models on its model page (read on its site at 19:55 UTC). It has no speed figure yet.

FOR CLAUDE SUBSCRIBERS

This week Anthropic is giving eligible paid plans one to three extra Fable resets, based on how long they have subscribed, and including Fable for up to 50% of weekly usage limits on Pro plans and standard Team seats.

HOW WE CAUGHT IT

AWS's public price list carried both models from 18:09:52 UTC, at the same $8 / $40, which we reported before the launch. A price list is not a served model, so the radar waited for the id: its catalog sweep saw claude-fable-5-2 at 18:52:10 UTC and fired at 18:53:06. It is now a row on /lead-time.

WHAT WE DO NOT KNOW

• How Fable 5.2 does on the other public boards. Artificial Analysis is the only one we have read so far.
• When Mythos 5.2 reaches organizations outside Anthropic's Life Sciences and Cyber Verification Programs.