Anthropic brought its most powerful AI model to the general public for the first time on June 9, 2026, with the launch of Claude Fable 5, the first publicly available version of its Mythos model. For anyone tracking frontier AI, this is the release that changes the reference point. Not an incremental update. A new tier entirely.
What Claude Fable 5 Actually Is
Fable 5 is the first model in a new "Mythos-class" tier that sits above the entire Opus line. It is available through the Claude API, Amazon Bedrock, and is included at no extra cost for Claude Pro, Max, Team, and Enterprise users during an introductory window from June 9 through June 22, 2026.
Fable 5 is the public face of a two-model release, paired with Claude Mythos 5, which is the same underlying model with safety classifiers lifted in restricted domains, available only to vetted Project Glasswing partners.
Key specifications at a glance:
- Priced at $10 per million input tokens and $50 per million output tokens, exactly double Opus 4.8.
- Supports up to 128k output tokens.
- API identifier: claude-fable-5, available across Claude API, Claude Code, AWS, Google Cloud, and Microsoft Foundry.
- Carries a 90% discount on input tokens via prompt caching.
Performance, Benchmarks, and Agentic Capabilities
Anthropic says Fable 5 exceeds every Claude model previously made generally available, featuring stronger performance across software engineering, knowledge work, vision, scientific research, and long-running tasks.
On SWE-Bench Pro, Fable 5 scores 80.3%, compared to Opus 4.8 at 69.2% and GPT-5.5 at 58.6%. It is also the first model to exceed 90% on Hex's analytical benchmark.
Additional official benchmarks place Fable 5 at 95.0% on SWE-bench Verified and 72.9% on CursorBench at max effort, with a leading position on FrontierCode in both Diamond and Main subsets.
What separates Fable 5 from earlier Claude releases is its autonomous reach. Third-party coverage confirms that Fable 5 can run agentic coding tasks for extended periods, described as days of work, a meaningful leap over prior Claude models.
Anthropic's Chief Product Officer, Mario Rodriguez, described it as "a clear step forward on agentic coding and prototyping."
Safety Architecture: How Anthropic Gated the Risk
This launch is as much about what Fable 5 does not do as what it does. In high-risk areas like cybersecurity, biology, chemistry, and distillation, the model blocks responses and falls back to Claude Opus 4.8.
Anthropic says those safeguards activate in less than 5% of sessions on average. When a reroute occurs, the user is notified.
Claude Mythos 5, the sibling model with some of those safeguards removed, stays restricted to Project Glasswing partners, the vetted cybersecurity and infrastructure firms that received access in April 2026.
Per the official API docs, both models run adaptive thinking permanently on, never return raw chain-of-thought, and introduce a new refusal-and-fallback billing model that no previous Claude release had used.
Purpose and Impact of Claude Fable 5
Claude Fable 5 is designed to perform complex coding, research, and long-running autonomous tasks with greater accuracy and reasoning.It raises the standard for AI models by advancing agentic AI, increasing competition with models from OpenAI, Google, and xAI, and pushing the industry toward more capable autonomous AI systems.
