Claude Opus 4.7 (xHigh) — benchmark results
Claude Opus 4.7 evaluated at the xhigh reasoning-effort setting. Provider: Anthropic. Released 2026-04-16. Access: API.
Unified ELO 1963 ± 65, rank #38 of 1776 rated models, from 26 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LisanBench | 1 | Mean Path Length / Current Maximum | 100 |
| OTIS Mock AIME 2024-25 | 97.8 | Accuracy (%) | 94.9 |
| FrontierMath - Tiers 1-3 | 43.79 | Accuracy (%, 290 problems) | 93.9 |
| SWE-Milestone | 41.29 | Milestone Score (%) | 90.5 |
| Epoch AI - ECI | 156.1 | ECI Score | 89.6 |
| FrontierMath - Tier 4 | 22.92 | Accuracy (%, 48 problems) | 88.7 |
| LiveBench | 77.1 | LiveBench average (self-reported) | 88.1 |
| MathArena - APEX 2025 | 40.62 | Accuracy (%) | 86.4 |
| ZeroBench | 14 | Score (%) | 86.4 |
| ProgramBench Almost | 4.5 | Almost (%) | 83.3 |
| SimpleQA Verified | 50.6 | Accuracy (%) | 74.2 |
| MathArena - ARXIV April | 58.54 | Accuracy (%) | 71.4 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-7-xhigh · How the rankings work · Data refreshed daily, snapshot 2026-07-22.