LLMs and models

Claude Fable 5.2

Quick explainer2 min readBeginner

Claude Fable 5.2 is Anthropic's top Claude model, released in November 2026 for hard reasoning, coding and long-running agent work.

In plain words

It is the same underlying model as Claude Mythos 5.2, but Fable has strong safeguards against cyber and biology misuse, so anyone can use it; Mythos needs an approved application. Anthropic says Fable 5.2 writes its output up to 30% faster than Fable 5.1 and is cheaper, at $8 for each million tokens it reads and $40 for each million it writes. An optional fast mode can make it up to 2.5 times quicker at double the price. It reads up to 1 million tokens at once.

A quick example

Imagine a team that runs most of its coding on Opus 5.5. When one large, vague project keeps stalling, they move just that project up to Fable 5.2 and leave the everyday tasks where they were.

Where you'll meet it

  • Open-ended technical projects, such as large coding jobs.
  • An advisor or coordinator that smaller Claude models hand hard problems to.

Related

Sources

  1. DocsClaude Fable 5.2 and Mythos 5.2, Anthropic · read 9 Nov 2026
  2. DocsModels overview, Claude Docs (Anthropic) · read 9 Nov 2026

Checked by WhatIsAI against the sources above · 9 Nov 2026. Spotted an error? Report it.