Claude Opus 4.1 (Non-reasoning): benchmark results
Claude Opus 4.1 evaluated with reasoning disabled. Provider: Anthropic. Released 2025-08-05. Access: API.
Aggregate Rating 1666 ± 29, rank #486 of 2257 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Generalization V1 (Lechmazur) | 1.69 | Avg Rank (lower is better) | 98.8 |
| UGI - Writing | 69.91 | Writing Score | 98 |
| UGI - Natural Intelligence | 58.74 | NatInt Score | 91.8 |
| Confabulation Leaderboard (Lechmazur) | 3.96 | Confabulation rate % (lower is better) | 91.5 |
| Elimination Game (Lechmazur) | 5.21 | TrueSkill μ | 84.7 |
| MeasureBench Synthetic | 13.3 | Accuracy (%) | 75 |
| Translation (Lechmazur) | 8.56 | Mean Score | 71.4 |
| NYT Connections Older Models | 21.7 | Score (%) | 63.4 |
| MeasureBench Real-world | 14.3 | Accuracy (%) | 58.3 |
| Chess Bench LLM | 303 | Lichess-Classical-calibrated Glicko-2 rating (native signed | 45.3 |
| Inverse Turing Bench | 57.99 | Accuracy (imported) | 42.9 |
| Step Game (Lechmazur) | 1.57 | TrueSkill μ | 40.5 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-1-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-17.