FAST TAKE · 2026-06-09 · ANTHROPIC CLAUDE FABLE 5
Claude Fable 5 — Anthropic's new ceiling, pulled three days after launch
Anthropic launched Claude Fable 5 (alongside the more capable, restricted Mythos 5) on June 9 — a "Mythos-class" model it calls the most capable it has ever made generally available, state-of-the-art on nearly every tested benchmark, priced at $10 / M input and $50 / M output (double Opus 4.8). On June 12, Anthropic suspended all access to both Fable 5 and Mythos 5 under a U.S. government export-control directive; as of this writing it's still down. It's hosted-only frontier, so it changes no local pick — and we are not adding it to the cost calculator while it's unavailable.
Verdict: Anthropic's most capable model ever — launched June 9 at $10/$50, then suspended June 12 under a U.S. export-control directive
The take
The facts, verified against Anthropic's own announcement: Fable 5 launched June 9, 2026 as a Mythos-class model "made safe for general use," with capabilities that "exceed those of any model we've ever made generally available" — strongest on long, complex software-engineering and knowledge work. API pricing is $10 / M input and $50 / M output, exactly twice Claude Opus 4.8 ($5 / $25) and the same per-token rate as Opus 4.8's fast mode. Then the dated update: on June 12, 2026 Anthropic posted "We are suspending access to Claude Fable 5 and Claude Mythos 5," attributing it to a U.S. government export-control directive to suspend all access.
Why it matters: two firsts in one week. The price is a new ceiling — $50 / M output is the most Anthropic has ever charged for a generally-available model, a clear bet that the value is in autonomous, multi-day work (Stripe is cited compressing "months of engineering into days" on a 50-million-line migration). And the suspension is the bigger story: a frontier lab's most capable public model pulled three days after launch by a government export-control order is a first, and a signal that the closed-frontier tier is now entangled with national-security policy in a way the local-weights world simply isn't.
Our call: nothing local changes, and we're holding it out of the cost calculator on purpose — you can't budget a plan around a model you can't access. This is exactly the asymmetry the site argues for: open weights you've already downloaded (Qwen, GLM, DeepSeek, Llama) can't be switched off by a directive. If Anthropic restores Fable 5 and the pricing holds, we'll add it to the calculator then. For now it's a dated marker of where the closed frontier — and its risks — are heading.
Where this fits
Models: DeepSeek V4-Pro · GLM-5.1 · Kimi K2.6
Hardware: Dual RTX 5090 · NVIDIA DGX Spark · Mac Studio M3 Ultra 96 GB
Sources
Next step
Try this in the planner→