Claude Opus 5.5 scores 72.4 on the Respan Index (#1), ahead of Claude Sonnet 5.5 at 68.5 (#6). Claude Opus 5.5 leads on 19 of the 26 benchmarks both report. Claude Sonnet 5.5 is 2.0x cheaper per token at list price.
* Estimated: no published scores in that category. How the Respan Index works
| Furniture Assembly (Epoch AI) | 83.3% | 75% |
| GPQA Diamond | 90.6% | 95.6% |
| LiveBench Agentic Coding (LiveBench) | 71.7% | 56.3% |
| LiveBench Coding (LiveBench) | 89.3% | 91.4% |
| LiveBench Data Analysis (LiveBench) | 80.3% | 78.6% |
| LiveBench Instruction Following (LiveBench) | 67% | 70.5% |
| LiveBench Language (LiveBench) | 86.3% | 83.4% |
| LiveBench Mathematics (LiveBench) | 97.1% | 96.7% |
| LiveBench Reasoning (LiveBench) | 92.2% | 91.6% |
| Mystery Game Puzzles (Epoch AI) | 71% | 65% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 100% |
| SAGE (Vals AI) | 45.8% | 51.8% |
| SimpleQA Verified | 72.2% | 46.5% |
| Terminal-Bench 4.0 (Vals AI) | 65.2% | 64.1% |
| LMArena Agent (LMArena) | 0.1382 | 0.1252 |
| LMArena Elo | 1504 | 1471 |
| LMArena WebDev (LMArena) | 1815 | 1786 |
| SimpleBench (SimpleBench) | 88.4% | 75.9% |
| CursorBench 4.0 | 57.8% | 55.5% |
| FrontierCode v1.1 (Main) | 54.4% | 46.2% |
| Humanity's Last Exam (with tools) | 67.7% | 64.5% |
| Terminal-Bench 4.0 | 66.4% | 70.6% |
Higher is better. Each score links to where it was published.