Claude Opus 5.5 scores 72.4 on the Respan Index (#1), ahead of Claude Fable 5.1 at 69.4 (#4). Claude Opus 5.5 leads on 19 of the 25 benchmarks both report. Claude Opus 5.5 is 2.5x cheaper per token at list price.
* Estimated: no published scores in that category. How the Respan Index works
| FrontierSWE V2 (Proximal Labs) |
| 56.3% |
| 62.3% |
| Furniture Assembly (Epoch AI) | 70% | 83.3% |
| LiveBench Agentic Coding (LiveBench) | 66.1% | 71.7% |
| LiveBench Coding (LiveBench) | 86.4% | 89.3% |
| LiveBench Data Analysis (LiveBench) | 80.3% | 80.3% |
| LiveBench Instruction Following (LiveBench) | 73% | 67% |
| LiveBench Language (LiveBench) | 89.5% | 86.3% |
| LiveBench Mathematics (LiveBench) | 97% | 97.1% |
| LiveBench Reasoning (LiveBench) | 91.7% | 92.2% |
| MirrorCode (Epoch AI) | 73.3% | 77.4% |
| Mystery Game Puzzles (Epoch AI) | 58% | 71% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 100% |
| SAGE (Vals AI) | 48.5% | 45.8% |
| SimpleQA Verified | 70.8% | 72.2% |
| Terminal-Bench 4.0 (Vals AI) | 58.1% | 65.2% |
| Vending-Bench 2 (Andon Labs) | 5421.56 | 9235.25 |
| LMArena Agent (LMArena) | 0.1431 | 0.1382 |
| LMArena Elo | 1501 | 1504 |
| LMArena WebDev (LMArena) | 1749 | 1815 |
| SimpleBench (SimpleBench) | 86.6% | 88.4% |
| Terminal-Bench Science 0.1 | 52.6% | 58.7% |
Higher is better. Each score links to where it was published.