Anthropic Launches Claude Fable 5.2 and Mythos 5.2 at Lower Prices

San Francisco, California — Artificial intelligence developer Anthropic has released Claude Fable 5.2 and Claude Mythos 5.2, upgraded versions of its top-tier models for hard technical work, and priced both at $8 per million input tokens and $40 per million output tokens.

Those rates are 20% below the $10 and $50 per million tokens that Anthropic charged for Fable 5.1. The company said the new versions also produce output up to 30% faster than Fable 5.1 and Mythos 5.1, and that testers found them clearer communicators than earlier Fable and Mythos releases.

An optional fast mode lifts token generation speed by as much as 2.5 times. It costs $16 per million input tokens and $80 per million output tokens, double the standard rate.

The two products run on one underlying model and differ only in their safeguards. Fable carries strict cybersecurity and biology protections against misuse, which lets Anthropic offer it to every user, while organizations doing beneficial cybersecurity or biomedical work can apply for Mythos through the recently expanded Life Sciences and Cyber Verification Programs.

During testing before release, Anthropic said, the models helped its engineers finish software migrations that would otherwise have been out of reach, designed three working electronic devices from scratch and solved two open problems in black hole physics. They also advanced well on robotics tests and produced a machine-checked version of the proof of Fermat’s Last Theorem within 30 hours, with gains reported across coding, reasoning and vision.

Anthropic is not pitching the pair as everyday tools. It advises starting most coding and knowledge tasks on Claude Opus 5.5, which it calls its strongest daily driver, and moving up to Fable or Mythos, or down to Claude Sonnet 5.5 or Claude Haiku 5.5, as the scope and complexity of a job require. Fable and Mythos are meant for the most ambitious, open-ended technical projects, it said.

In agentic coding, Fable 5.2 scored 49.1% on FrontierCode 2.0 at Xhigh effort, against 32.4% for Fable 5.1, 45.5% for Opus 5.5, 42.8% for rival model GPT-6 Astra and 41.1% for Gemini 4 Carbon at Medium effort. On FrontierSWE v2 the same five models scored 69.9%, 56.3%, 62.3%, 65.5% and 61.2%.

FrontierCode checks whether maintainers of open-source repositories would merge a model’s code changes. There, Fable 5.2 at High effort scored 45.3%, close to the 45.5% best that Opus 5.5 reached at Max effort while costing about half as much per task, and beat the 42.8% best of GPT-6 Astra at a similar cost. At Xhigh, its 49.1% is the top score of any released model Anthropic has tested.

Several of the early testers are trading and investment firms. Testers using Fable 5.2 for research and analysis told Anthropic it handles more ambiguous, data-heavy tasks than earlier models and shows better judgment on them, particularly where abstract reasoning and design are involved.

Lucas Baker, head of LLM R&D at Jump Trading, said: “Claude Fable 5.2 is incredible at playing every major role within a research system: defining questions, interpreting, coordinating, and of course finding and implementing the key ideas. Definitely one of the strongest on taste and judgment that we’ve seen. In an internal research automation benchmark, it achieved a full 10% lower error than the next best model.”

Alex Gerko, founder and CEO of XTX Markets, added: “In our quantitative research work, Claude Fable 5.2 delivered very strong improvements versus Fable 5.1, especially in the setup with multiple coordinating agents. Its maths research capabilities represent another meaningful advance, too.”

Anthropic’s Enterprise Frontier Safeguards (EFS) is now in private beta with customers including Bridgewater, Salesforce and Stripe. Enterprise customers can request access through their account team or an online form, eligible customers can keep using the new models with zero data retention before they switch on EFS monitoring, and other customers are subject to 30-day retention for safety monitoring.

Fable 5.2 is available now on all platforms, including Amazon Web Services, Google Cloud and Microsoft Azure, and developers can call it on the Claude Platform as claude-fable-5-2.