Anthropic launches Fable 5.2 and Mythos 5.2 with faster AI capabilities

Anthropic says its latest AI models offer faster responses and stronger coding, reasoning and research capabilities, while Fable 5.2 comes with lower pricing and safeguards

Anthropic on Monday announced the launch of its latest and most advanced artificial intelligence (AI) models, Fable 5.2 and Mythos 5.2. The company says these models offer significant improvements in coding, reasoning and visual tasks, along with clearer communication.

 

The upgraded models can generate outputs up to 30 per cent faster than their predecessors, Claude Fable 5.1 and Claude Mythos 5.1. Anthropic has also reduced Fable's pricing from $10 to $8 per million input tokens, making Fable 5.2 cheaper than its primary competitor, GPT-6 Astra.

 

What can Anthropic's Fable 5.2 and Mythos 5.2 do?

 

Anthropic said Fable 5.2 can handle more complex software engineering work, hardware-related tasks such as controlling robots, and long-running projects. It is also designed to coordinate more than 1,000 AI agents simultaneously.

 

The company said early testers found that the new models communicate more clearly than previous Fable and Mythos versions. They also showed better judgement on ambiguous, data-heavy research and analysis tasks, particularly those involving abstract reasoning and design.

 

Fable 5.2 can function as an adviser, with smaller models such as Haiku, Sonnet or Opus consulting it to solve difficult problems. It can also act as an orchestrator, delegating simpler tasks to smaller models.

 

Anthropic recommends that users begin with Claude Opus 5.5, which it describes as its strongest everyday model, for most coding and knowledge-based tasks. They can then move to Fable or Mythos for more complex work, or use Claude Sonnet 5.5 and Claude Haiku 5.5 for less demanding tasks.

 

Mythos 5.2 is being made available to select organisations and is described by Anthropic as its most capable model for cybersecurity engineering. The company says it can identify more vulnerabilities than earlier models and is better at reproducing and fixing them. Fable 5.2, meanwhile, has safeguards that block most cybersecurity-related tasks.

 

Safety testing

 

Anthropic said it conducted extensive pre-release safety testing on the model underlying both releases.

 

The company said high-stakes risks from misalignment remain low and that the model performed as well as, or better than, recent versions on most alignment measures. Fable 5.2's cybersecurity and biology safeguards are as robust as those of Fable 5.1, while producing fewer unnecessary blocks. Anthropic also said it had introduced its strongest protections yet against distillation attacks.

 

Where are Fable 5.2 and Mythos 5.2 available?

 

Claude Fable 5.2 is available across platforms, including Amazon Web Services, Google Cloud and Microsoft Azure. Anthropic is also offering eligible paid subscribers additional usage this week. Subscribers on plans and seats that include Fable models will receive one to three usage resets, depending on how long they have subscribed. Fable will also be included for up to 50 per cent of weekly usage limits for Pro plans and standard seats on the Team plan.

 

Cybersecurity professionals can apply for access to Mythos 5.2 through Anthropic's Cyber Verification Programme.

 

Fast mode for both models is available on Claude Code, the Claude Platform, Claude Models on Google Cloud (both Anthropic-managed and Google-managed) and Microsoft Foundry (Anthropic-hosted). Access to Mythos 5.2's fast mode is restricted to organisations participating in Anthropic's verification programmes.