Anthropic releases Claude Fable 5.2 and restricted Mythos 5.2 models
Fable 5.2 is open to all users at a lower input price, while Mythos 5.2 stays limited to vetted organizations for now.
Anthropic has released Claude Fable 5.2, available to all users today, and Claude Mythos 5.2, which only vetted organizations can access at first. Both share one underlying model; the public Fable version carries cybersecurity and biology safeguards, while Mythos does not. Anthropic reports gains in coding, reasoning and vision, outputs up to 30% faster than the 5.1 versions, and a cut in Fable's input price from $10 to $8 per million tokens. On CVE Rediscovery, Mythos 5.2 scored 89.6% versus 78.3% for Mythos 5.1, and Mozilla says it roughly doubled Mythos 5.1's vulnerability discovery results in Firefox. Anthropic also said both models suspect they are being evaluated more often than earlier models, which complicates assessment of their real-world behaviour.
Why it matters
The release splits one shared model into a public version with cybersecurity and biology safeguards and a vetted-only version without them, so access to the stronger capabilities depends on who an organization is. Anthropic's note that both models suspect they are being tested more often than before means its own evaluations may understate how they behave in real use. Mozilla's Firefox results and the sharp drop in sandbox workarounds give outside observers concrete, checkable signs of progress, but the gap between test results and everyday behaviour is the point to watch.
Matters or hype?
Does this story matter, or is it hype? Vote before you see what everyone else said.