GPT-5 scores 37.2 on the Respan Index (#54), ahead of Claude Sonnet 4.6 at 34.4 (#55). Claude Sonnet 4.6 leads on 8 of the 12 benchmarks both report. GPT-5 is 1.7x cheaper per token at list price. Claude Sonnet 4.6 has the larger context window (1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| 46.6% |
| 43.7% |
| SimpleQA Verified | 35.5% | 50.1% |
| SWE-bench Verified (Epoch AI) | 75.2% | 73.6% |
| WeirdML (Håvard Tveit Ihle) | 66.1% | 60.7% |
| LMArena Elo | 1458 | 1435 |
| LMArena Vision (LMArena) | 1275 | 1210 |
| LMArena WebDev (LMArena) | 1521 | 1417 |
| SWE-bench Verified | 79.6% | 74.9% |
Higher is better. Each score links to where it was published.