Across 9 shared benchmarks, Claude Opus 4 scores higher on 0 and GPT-6 Astra on 9. The widest gap is HLE, where GPT-6 Astra scores 54.7 against 12.3. GPT-6 Astra is the cheaper of the two on tracked API pricing ($10.00 against $15.00 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, direct), otherwise the lowest tracked offer.