AI
Anthropic Releases Claude Fable and Mythos 5.2 With Stronger Coding and Reasoning
The AI company unveiled upgraded versions of its flagship models, touting improvements in technical problem-solving and generation speed.
1/2
The Verge
Anthropic has unveiled two upgraded AI models—Claude Fable 5.2 and Claude Mythos 5.2—intended to tackle demanding computational and technical challenges with enhanced capabilities. According to the company, the new iterations deliver gains in coding, reasoning, and visual understanding while producing outputs more quickly than their predecessors.
What's New#
The refreshed models mark an incremental advance over the 5.1 releases that arrived in September. Claude Fable 5.2 is available to users immediately, while Claude Mythos 5.2 access is restricted to approved organizations for now.
Demonstrated Capabilities#
During testing before release, Anthropic reports the models have handled an ambitious range of tasks. The company notes achievements including handling complex software engineering migrations, designing three functional electronic circuits from first principles, addressing two long-standing theoretical physics questions about black holes, advancing robotics benchmarks, and completing a formal proof of Fermat's Last Theorem within 30 hours.
Focus on Complex Problem-Solving#
Fable 5.2 is positioned as particularly strong at handling intricate software engineering problems and enabling coordinated agent systems to work through extended projects. The model's improvements in reasoning and coding capacity aim to make it capable of wrestling with technical challenges that previously demanded human expertise or remained altogether unsolved.
More details are available in Anthropic's announcement.

