Claude Mythos 5 and Fable 5 have officially been unveiled by Anthropic, marking one of the company’s most significant AI releases to date. The launch introduces a new generation of advanced AI systems designed to deliver stronger performance in software development, research, analytics, and complex reasoning tasks while addressing growing concerns about AI safety and misuse.
Anthropic described Fable 5 as the first broadly available model from its new Mythos-class family. The company says the model offers major improvements in handling long and complex tasks, outperforming previous Claude models across several professional and enterprise use cases.
Fable 5 Brings Mythos-Level Capabilities to the Public
The centerpiece of the announcement is Claude Fable 5, a public-facing AI model built on the same foundation as Mythos 5. Anthropic says Fable 5 delivers exceptional performance in software engineering, knowledge work, and multimodal tasks involving images and visual analysis. The company claims its advantages become even more apparent during lengthy projects that require sustained reasoning and context retention.
Unlike earlier Mythos previews that were restricted to selected partners, Fable 5 is now accessible to a much broader audience. The move reflects Anthropic’s confidence that it has developed sufficient safeguards to make the technology available without exposing high-risk capabilities.
Why Anthropic Added Safety Guardrails
Anthropic previously acknowledged that Mythos-class models demonstrated unusually powerful capabilities in areas such as cybersecurity, prompting concerns about potential misuse. To address these risks, Fable 5 includes built-in safety mechanisms that automatically restrict responses involving sensitive topics like cybersecurity exploitation, biology, chemistry, and other potentially dangerous domains.
When users submit requests that fall into these high-risk categories, the system can redirect them to the more conservative Claude Opus 4.8 model instead of allowing direct access to Fable 5’s full capabilities. Anthropic says this layered approach allows the public to benefit from advanced AI performance while reducing the likelihood of harmful use.
Claude Mythos 5 Remains Restricted
Alongside Fable 5, Anthropic also introduced Claude Mythos 5, a version of the same model architecture that operates with fewer restrictions. However, Mythos 5 is not being released broadly. Instead, access is currently limited to trusted organizations participating in Anthropic’s Project Glasswing initiative and select partners in cybersecurity, infrastructure, and scientific research.
The company plans to expand access gradually through a trust-based program designed to ensure that organizations receiving Mythos 5 can responsibly manage its advanced capabilities.
Enhanced Performance for Enterprise Users
Anthropic says both models were built to handle increasingly demanding enterprise workloads. The new systems are designed to perform better in software development, technical analysis, financial research, strategic planning, and other complex professional applications. According to the company, performance gains become more noticeable as tasks grow longer and require deeper reasoning.
Read More:
OpenAI and Anthropic Enter Fierce Battle Over Free AI Coding Tools
The models also offer improved efficiency, allowing users to complete sophisticated tasks with fewer prompts and reduced interaction overhead. Anthropic believes this can help organizations achieve better productivity while lowering operational complexity.
New Pricing Structure
Anthropic has introduced premium pricing for the Mythos-class family. Reports indicate that Fable 5 and Mythos 5 are priced at approximately $10 per million input tokens and $50 per million output tokens, making them more expensive than earlier Opus models. However, the company argues that increased efficiency and stronger performance can offset higher token costs for many users.
Anthropic’s Biggest AI Launch Yet
The launch of Claude Mythos 5 and Fable 5 represents a major milestone in Anthropic’s AI roadmap. By combining powerful reasoning capabilities with extensive safety controls, the company aims to strike a balance between innovation and responsible deployment. While Fable 5 opens Mythos-class technology to the public, Mythos 5 remains reserved for carefully selected organizations, reflecting Anthropic’s cautious approach to managing frontier AI systems.
As competition in the AI industry continues to intensify, Anthropic’s latest models position the company among the leading developers of next-generation artificial intelligence, offering users access to more capable tools while maintaining a strong focus on safety and oversight.
