Claude Opus 5 scores 63.9 on the Respan Index (#7), ahead of GPT-6 Sol at 62.8 (#8). GPT-6 Sol leads on 12 of the 23 benchmarks both report. GPT-6 Sol is 2.5x cheaper per token at list price. GPT-6 Sol has the larger context window (1.1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| Furniture Assembly (Epoch AI) |
| 60.8% |
| 58.3% |
| GPQA Diamond | 93.9% | 94.3% |
| LiveBench Agentic Coding (LiveBench) | 65.2% | 52.9% |
| LiveBench Coding (LiveBench) | 81.4% | 81.8% |
| LiveBench Data Analysis (LiveBench) | 74.6% | 81.2% |
| LiveBench Instruction Following (LiveBench) | 63.8% | 68.6% |
| LiveBench Language (LiveBench) | 88.7% | 85.3% |
| LiveBench Mathematics (LiveBench) | 95.7% | 96.4% |
| LiveBench Reasoning (LiveBench) | 91.2% | 88.7% |
| Mystery Game Puzzles (Epoch AI) | 59% | 56% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 98.9% | 100% |
| SAGE (Vals AI) | 49.4% | 44.8% |
| SimpleQA Verified | 59.9% | 60.7% |
| Terminal-Bench 4.0 (Vals AI) | 53.5% | 44.4% |
| Vending-Bench 2 (Andon Labs) | 11181.87 | 14427.85 |
| LMArena Agent (LMArena) | 0.0867 | 0.0971 |
| LMArena Elo | 1490 | 1457 |
| LMArena WebDev (LMArena) | 1695 | 1689 |
| SimpleBench (SimpleBench) | 80.6% | 73.1% |
Higher is better. Each score links to where it was published.