Claude Opus 4.1
Anthropic·
LLMsflagshipcloud
Overview
Massive reasoning model for complex R&D tasks.
Capabilities and innovations
Deep researchScientific discoveryLong-horizon planning
Benchmarks
| Benchmark | Score | Source |
|---|---|---|
| GPQA Diamond | 88.5% | unsourced |
| Humanity’s Last Exam (no tools) | 38.9% | unsourced |
| SWE-bench Verified | 63.8% | unsourced |
| MMLU | 91.5% | unsourced |
| MMLU-Pro | 87.92% | unsourced |
- SWE-bench Pro: not reported by the vendor
- Terminal-Bench 2.x: benchmark did not exist at release
- MMMU-Pro: benchmark did not exist at release
Links
More from Anthropic
Data curated by AI Model Timeline. See the methodology for admission criteria, benchmark eras and source priority.