Claude Mythos vs Fable 5: What Is the Difference?

TL;DR
Claude Mythos 5 vs Fable 5 explained. They are the same underlying model with different access and safeguards, which is why both were suspended together.
Last updated: June 13, 2026
The One Thing Most People Get Wrong#
Mythos 5 and Fable 5 are not two different models. They are the same underlying model with two different wrappers around it.
That single fact explains almost everything confusing about the launch, the pricing, and why the US government suspended both at once on June 12, 2026.
The Actual Difference#
Anthropic split one model into two products based on safeguards and access, not raw intelligence.
Fable 5 is the generally available version. Anyone on the relevant plans can use it. It ships with Anthropic's full safety layer, the one many users complained was too aggressive, refusing legitimate security and research work. It also carries a required 30-day data retention policy that Anthropic added specifically so it could monitor for and mitigate jailbreaks.
Mythos 5 is the restricted-access version of the same model, available only to a small set of approved organizations, without that same broad safeguard layer. It is the higher-trust, lower-friction tier for vetted users.
So the mental model is simple:
- Same brain.
- Fable = public, heavily guard-railed, monitored.
- Mythos = restricted, fewer guard rails, approved orgs only.
Both sit in a new tier Anthropic positions above Opus in capability.
Why This Matters for the Suspension#
Because they are the same model, a concern about one is a concern about both. When the government's directive landed citing a jailbreak, there was no way to suspend Fable while keeping Mythos, or vice versa. The capability lives in the shared model, so Anthropic pulled both.
It also explains the irony at the center of the story. Anthropic built Mythos's restricted access and Fable's 30-day retention precisely to manage the risk of a model this capable. That posture is what put "national security concern" in the air around these models in the first place. We unpack that in Why the US Government Pulled Fable 5.
See Both for Yourself#
We covered the Mythos preview when it dropped and did a full hands-on with Fable 5. Both, side by side:
Which Should You Care About?#
For almost everyone, the answer is Fable, because Mythos was never available to you anyway. And right now neither is, because both are suspended. If you want the fuller story of the original gated model - the benchmarks, the access program, and the pricing - our Claude Mythos Preview explainer covers the April launch in depth.
If you are deciding what to actually build on today, the practical question is not Mythos vs Fable. It is which available model to use now that both are gone. We answer that in Best Claude Model Now That Fable 5 Is Disabled, and if you just want to know whether it is back yet, check Is Claude Fable 5 Down?. For the wider benchmark and pricing picture against OpenAI's frontier model, see Fable 5 vs GPT-5.5.
Official Sources#
| Source | Link |
|---|---|
| Anthropic Models Documentation | docs.anthropic.com/models |
| Anthropic Official Statement | anthropic.com/news |
| Claude API Pricing | anthropic.com/pricing |
| Claude Code Documentation | docs.anthropic.com/claude-code |
FAQ#
Are Mythos 5 and Fable 5 different models?#
No. They are the same underlying model wrapped in two different access and safeguard layers. Fable is the generally available, heavily guard-railed version; Mythos is the restricted-access version for approved organizations with fewer guard rails.
Why were both suspended at the same time?#
Because they share the same underlying model, a capability concern about one is a concern about both. Anthropic could not suspend Fable's public access while leaving Mythos live, since the capability that triggered the directive lives in the shared model itself.
Which one should most developers care about?#
Fable, since Mythos was never available outside a small set of approved organizations. Right now the more useful question is which currently available Claude model to use instead, not which of these two suspended tiers matters more.
Get the next comparison like this in your inbox
One email a week on Claude and the rest of the AI dev stack. Free.
Read next on Claude Code
Claude Mythos and Fable 5 Banned: The Export Controls That Shut Down Two Frontier Models
Claude Fable 5 and Mythos 5 were suspended for all users by a US export control directive. What it said, why it hit everyone, and what changed for developers.
6 min readClaude vs Claude Code: What the Difference Is and Which to Use (2026)
Claude vs Claude Code: one app for chat, docs and general work, one agent for your codebase. Same model family and shared plan limits, different jobs.
6 min readAnthropic Cuts Fable 5 Biology Fallbacks by 85%: What the Safeguard Tuning Means for Developers
Anthropic retuned Claude Fable 5 biology safeguards, cutting biology fallbacks about 85%. What changed and what still routes to Opus 5.
5 min readTechnical content at the intersection of AI and development. Building with AI agents, Claude Code, and modern dev tools - then showing you exactly how it works.








