Anthropic Launches Claude Fable 5: A Safe, Generally Available Mythos-Class Model

ClaudeClaude

· Updated

Anthropic launched Claude Fable 5, its first Mythos-class model made generally available with safeguards for broad use. This release brings frontier AI capabilities to a wider audience, demonstrating state-of-the-art performance across complex tasks like software engineering and scientific research.

Claude Fable 5, a new Mythos-class model (a tier of Claude models above Opus class in capability), is now generally available, offering capabilities that exceed Anthropic's previously released models. It achieves state-of-the-art performance on nearly all tested benchmarks, showing exceptional results in software engineering, knowledge work, scientific research, and vision, with its lead increasing on longer and more complex tasks.
Fallback Model
Claude Opus 4.8
Fallback Frequency
Less than 5% of sessions
Data Retention
30 days for Mythos-class models (for safety)
Mythos 5 Initial Access
Project Glasswing partners
Fable 5 Subscription Availability
Included until June 22, then usage credits
Safeguard Areas
Cybersecurity, biology and chemistry, distillation

This release balances capability with safety, offering Fable 5 for general use with safeguards and Claude Mythos 5—the same underlying model with lifted safeguards—for trusted partners. Classifiers detect requests related to cybersecurity, biology/chemistry, and distillation, falling back to Claude Opus 4.8 to prevent misuse, building on Project Glasswing.

Claude Fable 5 is available today via the Claude API and consumption-based Enterprise plans, with temporary inclusion on subscription plans until June 22. Priced at $10 per million input tokens and $50 per million output tokens, it makes advanced capabilities more accessible for various applications.

Claude Mythos 5 and Fable 5 benchmark performance across coding, reasoning, and specialized domain tasks versus competitors.
Claude
Claude
@claudeai
X

Introducing Claude Fable 5: a Mythos-class model that we’ve made safe for general use. Its capabilities exceed those of any model we’ve ever made generally available. https://t.co/2AvmEjHIX8

15kretweets105klikes
View on X

Still wondering? A few quick answers below.

Claude Fable 5 is Anthropic's new Mythos-class AI model, now generally available. It offers advanced capabilities that surpass previous models, excelling in complex tasks across software engineering, knowledge work, scientific research, and vision.

Claude Fable 5 includes new safeguards, such as classifiers that detect requests related to cybersecurity, biology and chemistry, and distillation. If a request is flagged, the model falls back to Claude Opus 4.8, ensuring safer interactions.

Claude Mythos 5 is a variant of the same underlying model as Fable 5, but with some safeguards lifted. It is initially restricted to Project Glasswing partners for defensive cybersecurity and will expand through a trusted access program.

Fable 5 is state-of-the-art on nearly all tested benchmarks. It shows exceptional performance in software engineering, knowledge work, scientific research, and vision, with its lead over other models increasing on longer and more complex tasks.

Claude Fable 5 is priced at $10 per million input tokens and $50 per million output tokens. This pricing is less than half the cost of Claude Mythos Preview, making its advanced capabilities more accessible.

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

Share this update