Across 16 shared benchmarks, Claude Fable 5 scores higher on 12 and Hy4 on 4. The widest gap is Program Bench, where Claude Fable 5 scores 76.8 against 17.5. Hy4 is the cheaper of the two on tracked API pricing ($0.83 against $10.00 per million input tokens).
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. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, tencent), otherwise the lowest tracked offer.