Anthropic has introduced Claude Fable 5, a new AI model that the company claims to be its most capable, generally available model to date. The company rolled out the new model on 9th June 2026; the new model is based on the same technology on which the company has developed Mythos 5, but it consists of additional safeguards designed for wider public access. According to the company, the model brings enhancements across coding, research, vision, reasoning, memory, and scientific applications.
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. pic.twitter.com/2AvmEjHIX8
— Claude (@claudeai) June 9, 2026
What Makes Claude Fable 5 Different
If you have been following Anthropic’s AI journey, you know that Mythos was the model everyone was talking about, but nobody could actually use. It was locked away due to serious cybersecurity concerns. Fable 5 changes that. Fable 5’s capabilities exceed those of any model Anthropic has ever made generally available, showing exceptional performance in software engineering, knowledge work, vision, scientific research, and many other areas. The longer and more complex the task, the bigger Fable 5’s lead over other models.
The model comes with a 1 million token context window, 128,000 maximum output tokens, and a knowledge cutoff date of January 2026. In simple terms, it can handle much longer conversations and much bigger tasks without losing track of what it is doing.
Fable 5 is not being presented as a separate smaller model. It shares the same class of capabilities as Mythos 5, just with stricter guardrails applied on top. Think of it as Mythos made safe enough for everyday users.
The Safety Guardrails Built In
This is where Anthropic has been careful. In high-risk areas like cybersecurity, biology, and chemistry, Fable 5 does not respond directly. Instead, it falls back to Claude Opus 4.8, a safer and more restricted model. Anthropic says this safety trigger will activate in less than five percent of sessions on average. So, for most day-to-day tasks, you will not even notice the switch happening.
The model includes safety classifiers that can refuse certain requests. When a refusal happens, it returns to a specific signal, and users are not charged if no output is generated. That is a thoughtful touch, making sure people do not pay for something the model refuses to do.
Businesses using Fable 5 will also face a mandatory 30-day data retention period for safety monitoring purposes. This is part of Anthropic’s broader effort to keep powerful AI in check while still making it accessible.
Who Can Use It and How
Fable 5 is available right now on the Claude API, Claude.ai chat interface, Claude Code for web, Claude Code CLI, and Claude Cowork. Developers can access it using the model string claude-fable-5.
Through June 22, Fable 5 is included in Pro, Max, Team, and seat-based Enterprise plans at no extra cost. From June 23 onwards, using it will require usage credits until Anthropic has enough capacity to bring it back as a standard subscription feature.
It is also available on AWS Bedrock and GitHub Copilot, which means developers working across different platforms can start experimenting with it straight away.
For anyone who has been waiting to get their hands on Anthropic’s most powerful AI, the wait is over. Just do not ask it to help you hack anything.
Syed Ziyauddin is a media and international relations enthusiast with a strong academic and professional foundation. He holds a Bachelor’s degree in Mass Media from Jamia Millia Islamia and a Master’s in International Relations (West Asia) from the same institution.
He has work with organizations like ANN Media, TV9 Bharatvarsh, NDTV and Centre for Discourse, Fusion, and Analysis (CDFA) his core interest includes Tech, Auto and global affairs.
Tweets @ZiyaIbnHameed
