Anthropic · model
Claude Opus 4.5 (non-reasoning) benchmark scores
As of 2026-10-07, Claude Opus 4.5 (non-reasoning) (Anthropic) has measured scores on 1 benchmark.
| Benchmark | Score | Source |
|---|---|---|
| MMLU-Pro | 88.9% | measured |
As of 2026-10-07, Claude Opus 4.5 (non-reasoning) (Anthropic) has measured scores on 1 benchmark.
| Benchmark | Score | Source |
|---|---|---|
| MMLU-Pro | 88.9% | measured |