Claude Fable 5 outpaces GPT-5.5 by 13 points on FrontierMath's toughest problems

Anthropic's Claude Fable 5 achieved 88% accuracy on FrontierMath's most challenging problems, significantly surpassing OpenAI's GPT-5.5, which scored 75%. This represents a substantial improvement from Opus 4.5, which had an accuracy below 10% in early 2026. The advancements in Claude Fable 5 indicate a rapid progression in AI's mathematical capabilities, which is crucial for practitioners developing applications that require robust problem-solving and reasoning skills.