Across 15 shared benchmarks, Agnes 2.5 Pro Beta scores higher on 3 and Claude Opus 4.8 on 12. The widest gap is OmniScience Accuracy, where Claude Opus 4.8 scores 48.8 against 16.8. Agnes 2.5 Pro Beta is the cheaper of the two on tracked API pricing ($0.10 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 (direct, anthropic-official), otherwise the lowest tracked offer.