Anthropic releases Fable 5.2 and Claude Mythos 5.2
Anthropic has launched Fable 5.2, a frontier model designed for complex engineering and creative tasks, alongside a restricted version for cybersecurity and life sciences. The release positions the new model as a faster, more capable alternative to previous versions for autonomous, long-running workloads.
Anthropic has released Fable 5.2, positioning it as a distinct frontier model separate from Opus 5.5, which remains the recommended tool for daily coding and knowledge tasks. The company also launched Claude Mythos 5.2, which utilizes the same underlying model but is restricted for cybersecurity and life sciences purposes. Fable 5.2 is highlighted for its capabilities in handling scope, complexity, and creativity, including solving two open problems in the physics of black holes. The model features cyber and bio guardrails, as well as anti-distillation safeguards, and is available via API and cloud platforms including AWS, Google Cloud, and Microsoft Azure.
Hands-on testing revealed that Fable 5.2 is highly effective for stubborn bugs, performance optimization, and long-running engineering tasks, including large migrations and operating system builds. The model demonstrated superior geometric reasoning by successfully generating assembly instructions for flat-pack furniture from low-resolution images with Chinese labels, a task that previous models failed to complete. In creative coding, Fable 5.2 generated a fully functional 3D rendered typewriter application with multiple scenes, audio integration, and printing capabilities in a single shot. However, the model performed poorly in front-end web design and image generation, often requiring pairing with Opus 5.5 or OpenAI's image gen 2.5 for aesthetic quality. Anthropic notes that the model outpaces other frontier models at their highest effort levels on benchmarks such as Frontier Code, Frontier Suite, and Automation Bench.