Anthropic
Claude 3 Opus benchmark results
9 results on 8 benchmarks from 4 sources.
Closed Released: Feb 29, 2024 Updated Oct 22, 2026
Also listed as: claude-3-opus-20240229 · Claude 3 Opus (20240229) · SWE-agent + Claude 3 Opus · RAG + Claude 3 Opus · Claude-3-Opus
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1322 | 256 | Overall, style control on; 194909 votes; 95% CI 1319–1325“claude-3-opus-20240229” | Official leaderboard | Oct 8, 2026 |
| LMArena Vision | 1062 | 136 | Overall, style control on; 15565 votes; 95% CI 1051–1072“claude-3-opus-20240229” | Official leaderboard | Oct 2, 2026 |
| GPQA Diamond | 47.2% | 220 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
| MMLU | 84.6% | 5 | Exact match; 5-shot | Official leaderboard via Epoch AI | Feb 29, 2024 |
| MMLU-Pro | 68.5% | 107 | Overall accuracy; data source: TIGER-Lab“Claude-3-Opus” | Official leaderboard | Mar 11, 2026 |
| SWE-bench Verified | 15.8% | 102 | agent: SWE-agent“claude-3-opus-20240229” | Official leaderboard | Apr 2, 2024 |
| SWE-bench Verified | 7.0% | 106 | agent: RAG baseline“claude-3-opus-20240229” | Official leaderboard | Apr 2, 2024 |
| AIME (OTIS Mock AIME 2024–2025) | 4.7% | 234 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Feb 25, 2025 |
| MATH Level 5 | 37.5% | 74 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
FAQ
Where do these Claude 3 Opus scores come from?
From Epoch AI Benchmarking Hub, LMArena leaderboard dataset, MMLU-Pro leaderboard (TIGER-Lab), and SWE-bench leaderboards (swebench.com). Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.