Canonical saga
Every verified milestone in Anthropic’s Mythos-class story, newest first — from the Glasswing preview to the launch of Fable 5 and Mythos 5 and the export-control suspension that followed. Curated anchors are augmented automatically as new reporting lands. Definitions → mythoslm.ai.
Release coverage from The Decoder: "Anthropic says its unreleased model is both more capable and better aligned than Mythos 5.1." See the original report for full details via the source link below.
Release coverage from 36Kr: "Fable 5.5 Surprise Launch Confirmed for Tomorrow – Google & OpenAI Fully Gear Up to Meet the New Challenge." See the original report for full details via the source link below.
The Anthropic status board records an active incident — "We’ve suspended access to Claude Mythos 5 and Claude Fable 5" — affecting claude.ai, the Claude API, and Claude Code, and links to the access statement.
Anthropic issues a statement on a US government export-control directive suspending all access to Fable 5 and Mythos 5. The company says it intends to restore Fable 5 to subscription plans when capacity and clearance allow.
Anthropic releases Fable 5 and Mythos 5 — the same underlying model, differentiated by safeguards. Fable 5 ships with safety classifiers that reroute flagged requests to Claude Opus 4.8; Mythos 5 lifts those classifiers under Glasswing-only access.
Fable 5 becomes available on the Claude API, the Claude Platform on AWS, Amazon Bedrock, Google Cloud, and Microsoft Foundry, with 1M-token context and up to 128k output. It is free for Pro, Max, Team, and Enterprise plans through June 22, with usage-based pricing after June 23.
The first Mythos-class model — a tier above Opus — reaches roughly 50 selected organizations in cyberdefense and critical infrastructure, including the US government, under the limited-access Project Glasswing program.