
Anthropic's Fable 5 topped GPT 5.5 on every major benchmark, scoring 80.3% to 58.6% on SWE-Bench Pro, before the US government ordered it shut down on June 12. GPT 5.5 is now the strongest available model, not because it improved but because its competitor was removed.
Tap to vote and see what everyone thinks.
Summary by ByteBrief
Claude Fable 5 beats GPT-5.5 by 13 points on FrontierMath