THU, NOVEMBER 12, 2026
Independent · In‑Depth · Practitioner‑Tested
Home › Large Language Models › Claude Fable 5.2
Large Language Models

Claude Fable 5.2

by Anthropic  ·  Paid

Anthropic flagship model for demanding technical work, released 9 November 2026. $8 input and $40 output per million tokens. Shares an underlying model with Claude Mythos 5.2, which is restricted to vetted organisations.

Visit Website ↗
New
☆☆☆☆☆No ratings yet
Avg Score
1 Articles
2 Compared

About Claude Fable 5.2

Claude Fable 5.2 is Anthropic flagship model for demanding technical work, released on 9 November 2026. API pricing is $8 per million input tokens and $40 per million output tokens, with an optional faster tier at $16 and $80.

Fable shares its underlying model with Claude Mythos 5.2. The difference is access and safeguards rather than capability: Fable is available to everyone with stronger cybersecurity and biology safeguards, while Mythos goes only to vetted organisations through a verification programme.

Published results include 69.9% on FrontierSWE v2, 49.1% on FrontierCode 2.0 at Xhigh effort, 74.2% on Humanity's Last Exam with tools, 89.3% on OSWorld 2.1 for computer use and 73.4% on Chartography for chart reading. Anthropic does not publish a context window figure on the announcement page.

✅ Key Features

  • $8 input / $40 output per million tokens
  • Fast tier at $16 input / $80 output
  • Image input supported
  • Effort levels including High and Xhigh
  • Shares a base model with Claude Mythos 5.2
  • API model ID claude-fable-5-2
  • Closed weights, API only

🧩 Specs

Platforms
APIWeb

🎯 Best For & Tags

Best For
Agentic codingComputer useScientific researchHardware engineeringKnowledge work automation
Tags
AnthropicClaudeLLMCoding AI2026AI agents

⭐ User Reviews

Write a Review →
⭐

No reviews yet. Be the first to review Claude Fable 5.2.

Write Review →