Across 14 shared benchmarks, Claude Fable 5 scores higher on 13 and K-EXAONE 2.0 on 1. The widest gap is OmniScience Accuracy, where Claude Fable 5 scores 65.3 against 13.1.
Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.