Anthropic releases Claude Fable 5, first public Mythos-class model at $10/$50 per million tokens
Anthropic has released Claude Fable 5, marking the first broad release from its Mythos class of AI models. The company previously deemed this model family too dangerous for public release due to exceptional cybersecurity capabilities, but new safeguards that block responses in high-risk areas now make it available at $10 per million input tokens and $50 per million output tokens.
Claude Fable 5 — Quick Specs
Anthropic releases Claude Fable 5, first public Mythos-class model at $10/$50 per million tokens
Anthropic has released Claude Fable 5, the first broadly available model from its Mythos class, which the company previously said was too dangerous to release publicly due to exceptional cybersecurity capabilities.
Pricing and availability
Fable 5 is priced at $10 per million input tokens and $50 per million output tokens — double the rates of Claude Opus 4.8, Anthropic's previous flagship model. The pricing is half what users currently pay for Mythos Preview access.
Safeguards and fallback mechanism
According to Anthropic, the release is "made possible by new safeguards that block responses in specific high-risk areas." When Fable 5 determines a query falls into a restricted category, it falls back to Claude Opus 4.8, which Anthropic described as particularly strong on "honesty" when launched last month.
The company singled out cybersecurity and biology as domains where safeguards may trigger. In testing, 95 percent of Fable sessions ran entirely on Fable responses without requiring fallback to Opus 4.8.
Performance claims
Anthropic claims Fable 5 "shows exceptional performance in software engineering, knowledge work, and vision," with its lead over other models growing as tasks become longer and more complex. The company describes it as "the most powerful model it has ever made widely available."
Claude Mythos 5: Limited access variant
Anthropic is also releasing Claude Mythos 5, which the company says is "the same underlying model as Fable 5, but with the safeguards lifted in some areas." Access to Mythos 5 appears limited to organizations in Anthropic's Project Glasswing initiative, which the company describes as "not entirely watertight." Anthropic said it plans to "expand access over time through a more systematic trusted-access program."
Unanswered questions
Anthropic did not explain how either model relates to Claude Mythos Preview or why the models are numbered "5" when there appear to be no previously released Mythos or Fable models. The company did not respond to requests for comment on these questions.
What this means
This release represents a significant shift in Anthropic's approach to powerful model deployment. Rather than withholding capable models entirely, the company is attempting to balance broad access with safety through automated guardrails and selective fallback to less capable systems. The dual-release strategy — a safeguarded public version and a less-restricted private version — suggests Anthropic is testing how to commercialize highly capable models while managing risks the company itself has identified as serious. The missing context about previous Mythos/Fable versions and the relationship to Mythos Preview raises questions about Anthropic's model versioning and communication strategy.
Related Articles
Anthropic's Claude Fable 5.1 Reportedly Solves 1653 Royalist Cipher in 44 Minutes
According to testing firm Vals AI, Anthropic's Claude Fable 5.1 independently identified and solved the 'Cyphral Distich,' a 1653 numeric cipher by Sir Thomas Urquhart that had defeated other frontier models. The AI decoded a hidden pro-royalist message by mapping each number to a word in Urquhart's original text.
OpenAI Launches GPT-6 Astra With Half the Message Allowance of GPT-5.6 Sol
OpenAI has begun rolling out GPT-6 Astra to top-tier ChatGPT plans, the API, Azure, and AWS Bedrock. The model delivers roughly half the usage allowance of GPT-5.6 Sol across comparable plans, with Plus and Business users gaining access in the coming days.
OpenAI's GPT-6 Astra Reportedly Automates AI Engineering Tasks at Under $6 an Hour, According to Latent Space Testing
A Latent Space report describes GPT-6 Astra, a new OpenAI model the blog says can autonomously handle AI engineering tasks—training models, labeling data, deploying systems—at an estimated cost of under $6 per hour. The claims, including 97.6% on FrontierMath and 99.9% on ARC-AGI-3, come from independent blog testing rather than an official OpenAI announcement.
Alibaba Releases Qwen-Drive 1.0, an Open Driving Model That Explains Its Own Decisions
Alibaba has released Qwen-Drive 1.0, a driving model built on Qwen3.5-4B that handles spatial perception, route planning, and cockpit dialogue in a single system. Reinforcement learning cut the rate of off-road driving errors in simulation from 24 percent to 12 percent, though the model's stated reasoning doesn't always match its actual maneuvers.
Comments
Loading...