article --summary
The summary
This piece walks through Anthropic's new Fable 5.2 and Mythos 5.2 models, covering their availability across cloud platforms, the restricted access to Mythos, and how the two differ in safety safeguards. It also explains the price cut from $10/$50 to $8/$40 per million tokens, the new Fast mode, and how the models compare with Opus 5.5 and OpenAI's GPT-6 Astra.
The article highlights benchmark results where Fable 5.2 leads on FrontierCode 2.0 and FrontierSWE v2, while noting places where Gemini 4 Carbon or Opus still come out ahead. It also covers Anthropic's claims about fewer safety blocks, enterprise safeguards in private beta, and experiments in physics, formal mathematics, and hardware design.
Readers will get a sense of what changed in this release and which claims come from the company itself rather than independent testing.
Editorial note: This is a summary of an article first published by The New Stack. The full piece is on their site.