Many MindedAnthropic ships Fable 5.2 and Mythos 5.2, with cheaper strong results
On November 9, 2026, Anthropic announced Claude Fable 5.2 and Claude Mythos 5.2. They are the same underlying model. Fable carries cybersecurity and biology safeguards and is available to all users. Organizations doing cybersecurity or biomedical research can apply for Mythos through the expanded Life Sciences and Cyber Verification Programs.
Pricing is $8 per million input tokens and $40 per million output tokens. Output generation is up to 30% faster than Fable 5.1 and Mythos 5.1. An optional fast mode speeds generation up to 2.5x, at $16 per million input and $80 per million output tokens.
Anthropic's own benchmarks carry the headline. On FrontierCode 2.0 at Xhigh effort, Fable 5.2 scored 49.1%, against 32.4% for Fable 5.1, 45.5% for Opus 5.5, 42.8% for GPT-6 Astra and 41.1% for Gemini 4 Carbon. At High effort it scored 45.3% versus Opus 5.5's best 45.5%, at roughly half the cost per task. On Chartography at Medium effort it scored 66.4%, ahead of Opus 5.5 at Max effort (64.4%), for under one-tenth of the cost per task. Anthropic also says over 90% of changes merged into its core codebase are written by Claude with human guidance, and one engineer used over 1,000 subagents across eight days to produce a bootable prototype operating system.
For abundance, the point is cost per task. When a model matches the previous best at half the cost, or beats it at a tenth, expert-level coding and analysis work gets cheaper to access. Splitting access by use also shows how capability gets gated where misuse risk is highest.
Caveats: the benchmark results and prerelease achievements are Anthropic's own claims. CyScenarioBench uses different tasks for each lab, and RoboDojo Sim was run internally. Scores are with safeguards on unless noted, and Anthropic's belief that high-stakes misalignment risks remain low is its own assessment. Anthropic still recommends Opus 5.5 as the default for most coding and knowledge tasks.
Source: Google News