Claude Opus 5 (Max): benchmark results

Provider: Anthropic. Released 2026-07-24. Access: API.

Aggregate Rating 1909 ± 6, rank #27 of 2169 rated models, from 402 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
ASI-Bench (Claude Code)50.29Scientific Score (0-100)100
ASI-Bench (Claude Code) - B1 Full Guidance72.29Scientific Score (0-100)100
ASI-Bench (Claude Code) - B2 Method Hints45.8Scientific Score (0-100)100
ASI-Bench (Claude Code) - B3 Goal Only40.7Scientific Score (0-100)100
ASI-Bench (Claude Code) - B4 Goal + Distractors42.39Scientific Score (0-100)100
BenchBench-Protocol (Best of 10)75.3Expected best-of-10 normalized rubric score (%; per task the100
BenchBench-Protocol - Cell Culture and Cell-Based Assays58.4Normalized rubric score (%; weighted rubric elements met by 100
BenchBench-Protocol - Genomics, Transcriptomics, and Sequencing56.1Normalized rubric score (%; weighted rubric elements met by 100
BenchBench-Protocol - Microbiology and Microbial Methods58.3Normalized rubric score (%; weighted rubric elements met by 100
BenchBench-Protocol - Molecular Biology and Cloning64Normalized rubric score (%; weighted rubric elements met by 100
BenchBench-Protocol - Protein Biochemistry and Purification58.1Normalized rubric score (%; weighted rubric elements met by 100
BenchBench-Protocol - Staining, Imaging, and Cytometry58.9Normalized rubric score (%; weighted rubric elements met by 100

Interactive version: theaggregate.ai/model?slug=claude-opus-5-max · How It Works · Data refreshed daily, snapshot 2026-10-23.