Across 18 shared benchmarks, Claude Opus 4.8 scores higher on 10 and Hy4 on 7, with 1 level. The widest gap is Program Bench, where Claude Opus 4.8 scores 71.9 against 17.5. Hy4 is the cheaper of the two on tracked API pricing ($0.83 against $5.00 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 (anthropic-official, tencent), otherwise the lowest tracked offer.