Claude Fable 5: Anthropic's Most Powerful Model, Its Ban, and Return
Fable 5 launched on June 9, 2026 as the first Mythos-class model available to the public. Three days later, the US government suspended it. On July 1, it came back. Here is the full story.
On June 9, 2026, Anthropic released Claude Fable 5 and Claude Mythos 5, the most capable models the company has ever made generally available. Fable 5 was the first in the Mythos class — Claude's highest tier, designed for complex reasoning, long-horizon agentic work, and advanced software engineering. Anthropic described it as a qualitative step beyond Opus 4.8.
The numbers backed the claim. On SWE-Bench Pro, a benchmark measuring resolution of real-world software engineering tasks, Fable 5 scored 80.3%. Opus 4.8 reached 69.2%, GPT-5.5 58.6%, and Gemini 3.1 Pro 54.2%. Stripe reported that Fable 5 compressed months of engineering work into days, completing a codebase migration that would have taken a full team two months. The model shipped with a one-million-token context window by default and up to 128,000 output tokens per request.
Built for autonomous work
The Mythos class is not simply a more powerful Opus. Anthropic designed it specifically for agentic tasks: long reasoning chains, robust tool use, reliable execution without constant human oversight. Where Opus 4.8 stops and asks for confirmation, Fable 5 keeps searching for solutions. Where Opus might miss a complex design flaw, Fable 5 catches it. The company described the model as capable of working at senior research scientist grade: choosing directions, allocating resources, and producing original first-principles outputs.
Mythos 5, the restricted sibling, launched in parallel but only for a small group of cyberdefenders and infrastructure providers, through Project Glasswing in collaboration with the US government. The general public accessed Fable 5; Mythos 5 remained outside the open market.
The unexpected ban
Three days after launch, on June 12, Commerce Secretary Howard Lutnick sent a letter to Anthropic CEO Dario Amodei ordering the immediate suspension of all access by any foreign national to Fable 5 and Mythos 5, whether inside or outside the United States. The stated reason: a jailbreak technique identified by Amazon researchers that allowed the model to generate information about security vulnerabilities and, in at least one case, code to exploit them.
Anthropic could not comply with the directive partially — no real-time mechanism existed to verify users' nationality — and chose to deactivate both models globally. Hundreds of millions of users and thousands of businesses that had integrated them into critical operations lost access overnight. The company published a statement openly disagreeing with the decision: it considered the jailbreak narrow, argued that it did not unlock general Mythos-level capabilities, and maintained it did not justify recalling a commercial model deployed worldwide. Nonetheless, it complied.
The return, with four commitments
On June 30, Lutnick notified Anthropic that the export controls were being lifted. On July 1, 2026, Fable 5 became available globally again on Claude.ai, Claude Code, the API, and Claude Cowork. Mythos 5, however, was reintroduced only for approved US organizations.
The reactivation came with four commitments from Anthropic to the government: proactively detect and address security risks in the models, collaborate on protocols and standards for future releases, report any malicious activity detected, and deploy a new security classifier that blocks the bypass technique in over 99% of cases.
Current situation for businesses
For Pro, Max, Team, and select Enterprise plans, Fable 5 is included up to 50% of the weekly usage limit through July 7. After that date, access is billed on usage credits. For standard Enterprise seats, there is no included allowance: access requires explicitly enabled credits. The new classifier may reroute some sensitive code generation requests to Opus 4.8; Anthropic has stated it will tune the classifier over coming weeks.
What has become clear is that frontier AI models can go from available to blocked within hours. For businesses that depend on a single API for critical operations, the Fable 5 episode is a warning about the need to manage technological dependency more carefully.
Sources
Anthropic — Claude Fable 5 and Mythos 5 announcement — https://www.anthropic.com/news/claude-fable-5-mythos-5
Anthropic — Statement on the US government directive — https://www.anthropic.com/news/fable-mythos-access
Anthropic — Redeploying Claude Fable 5 — https://www.anthropic.com/news/redeploying-fable-5
Forbes — Anthropic Disabled Fable 5 and Mythos 5 After A U.S. Export-Control Order — https://www.forbes.com/sites/anishasircar/2026/06/16/anthropic-disabled-fable-5-and-mythos-5-after-a-us-export-control-order-heres-what-happened/
News
Need our services?
Tell us what you need and we'll reply as soon as possible.