Claude Fable 5.2 and Mythos 5.2

Anthropic released Claude Fable 5.2 for all eligible users and Mythos 5.2 for authorized organizations, faster and cheaper than the 5.1 models.

Fable 5.2 and Mythos 5.2 share one underlying model with different guardrails: Fable has stronger cybersecurity and biology safeguards and is broadly available (Claude apps, AWS, Google Cloud and Microsoft Azure), while Mythos is limited to authorized organizations. Anthropic says both generate up to 30% faster than 5.1 and cost $8 per million input tokens and $40 per million output tokens. In Anthropic's own tests Fable 5.2 scored 49.1% on FrontierCode 2.0 at XHigh effort (Opus 5.5: 45.5%, GPT-6 Astra: 42.8%) and more than doubled Fable 5.1 and GPT-6 Astra on its internal RoboDojo Sim robotics benchmark. Anthropic also reports less misleading behavior and fewer sandbox-escape attempts by Mythos 5.2 in deliberately weakened test setups (about 3.1% of scenarios vs 21% for Mythos 5.1). These are company figures.

Status

Released

Date

Nov 9, 2026

Maker

Anthropic

Type

AI model

Price

$8 / $40 per million input / output tokens

Categories

AI

Sources

• Anthropic (official)
• Android Authority
• The Verge

Launch dates can change. Check the sources for the latest. Last reviewed Nov 10, 2026.