Claude Opus 4.5 (High): benchmark results
Claude Opus 4.5 evaluated at the high reasoning-effort setting. Provider: Anthropic. Released 2025-11-24. Access: API.
Aggregate Rating 1745 ± 16, rank #245 of 2167 rated models, from 61 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AsgardBench | 76.5 | Task success rate (%) with image observations (the previous | 100 |
| AsgardBench (Text-Only) | 36.1 | Task success rate (%) with text-only observations (no images | 100 |
| Olfactory Perception Benchmark - Rate-All-That-Apply | 42.2 | Score (%) on rate-all-that-apply descriptor profiling (selec | 100 |
| SWE-bench Verified | 76.8 | Resolved (%) | 100 |
| Sonar LLM Leaderboard - Java - Code Smell Density | 13.79 | Code smells per 1,000 lines of code (lower is better) | 100 |
| Tau-Bench Airline | 84 | Pass@1 (%) | 100 |
| NonoBench (v1.0) | 60 | Standard accuracy (%), correct / non-failed attempts; versio | 97.3 |
| IKP - T3 | 99.45 | Accuracy (%) | 95.4 |
| Olfactory Perception Benchmark - Odor Classification | 92 | Score (%) on odor classification (odorous or odorless, 175 m | 95 |
| IKP - T2 | 100 | Accuracy (%) | 92.6 |
| Sonar LLM Leaderboard - Java - Pass Rate | 83.63 | Passing tests (%) | 90.3 |
| ALFWorld | 1 | ALFWorld normalized score | 88.9 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-5-high · How It Works · Data refreshed daily, snapshot 2026-10-21.