Claude Opus 4.7 (Max): benchmark results
Claude Opus 4.7 evaluated at the max reasoning-effort setting. Provider: Anthropic. Released 2026-04-16. Access: API.
Aggregate Rating 1811 ± 10, rank #116 of 2168 rated models, from 133 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BackendForge (Base Oracle) | 58.93 | Task success rate (%; base oracle of 7,250 items written bef | 100 |
| HWE-Bench (Claude Code) | 74.6 | Resolved (%) | 100 |
| SafePyramid - Exact Match | 35 | Rule matching rate (%) over all 3,000 SafePyramid cases: per | 100 |
| SafePyramid - L0 Rule Understanding | 73.8 | Rule matching rate (%) on the 1,000 L0 cases (independent de | 100 |
| SuperCLUE-Writing - Content Quality | 90.77 | Score | 100 |
| SuperCLUE-Writing - Overall | 87.97 | Score | 100 |
| Vals AI SAGE | 56.1 | Accuracy (%) | 100 |
| AA-Omniscience Index - Software Engineering (SWE) - Swift | 84 | Omniscience Index | 97.3 |
| AA-Omniscience Index - Software Engineering (SWE) - Dart | 70 | Omniscience Index | 96.3 |
| Korean CSAT 2026 (Easy Mode) - Society and Culture | 47 | Points (out of 50) | 96.3 |
| APEX-Agents | 50.6 | Mean Score (ReAct) (imported) | 95.5 |
| AA Terminal-Bench Hard | 51.52 | Accuracy (%) | 95.4 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-7-max · How It Works · Data refreshed daily, snapshot 2026-10-20.