Anthropic releases Claude Fable 5.2 and Mythos 5.2 at $8 and $40 per million tokens
Anthropic says the two models are one underlying model, with Fable open to all users and Mythos limited to vetted cyber and biomedical organizations.
What happened
Anthropic has released two upgraded AI models, Claude Fable 5.2 and Claude Mythos 5.2. They are the same underlying model. Fable has extra safeguards against misuse in hacking and biology, so everyone can use it. Mythos is for approved organizations.
Why it matters to you
Anthropic says the models can do large technical jobs, such as rewriting software and designing electronics, with little human help. That could change how engineering and research teams work.
The takeaway
Anthropic has a stronger, cheaper model family, but its most sensitive abilities stay behind an application process.
What it means for you
Anthropic recommends starting with Opus 5.5 for most coding and moving up to Fable or Mythos for ambitious work, and says Fable 5.2 can act as an Advisor or Orchestrator for smaller models.
Anthropic says Claude Enterprise customers can use Claude Security, which now includes Mythos 5.2, to scan codebases for vulnerabilities without joining the Cyber Verification Program.
Anthropic lists lower per-token prices than earlier versions and says Fable 5.2 at High effort matched Opus 5.5 at Max effort on FrontierCode 2.0 for roughly half the cost per task.