Across 25 shared benchmarks, GPT-6 Astra scores higher on 16 and Claude Opus 5 on 9. The widest gap is arc_agi_3, where GPT-6 Astra scores 99.9 against 30.2. Claude Opus 5 is the cheaper of the two on tracked API pricing ($5.00 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, anthropic-official), otherwise the lowest tracked offer.