GPT-5.5 scores 52.5 on the Respan Index (#15), ahead of Grok 4.6 at 51.0 (#19). GPT-5.5 leads on 15 of the 25 benchmarks both report. Grok 4.6 is 3.8x cheaper per token at list price. GPT-5.5 has the larger context window (1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| 72.5% |
| 31.7% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 85.3% | 66% |
| Furniture Assembly (Epoch AI) | 44.2% | 40% |
| GPQA Diamond | 90.7% | 94% |
| LiveBench Agentic Coding (LiveBench) | 54% | 57% |
| LiveBench Coding (LiveBench) | 82.1% | 76.8% |
| LiveBench Data Analysis (LiveBench) | 81.6% | 73.9% |
| LiveBench Instruction Following (LiveBench) | 70.7% | 71.9% |
| LiveBench Language (LiveBench) | 87.4% | 83.7% |
| LiveBench Mathematics (LiveBench) | 95.9% | 92.6% |
| LiveBench Reasoning (LiveBench) | 89.7% | 90.5% |
| Mystery Game Puzzles (Epoch AI) | 56% | 34% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 84.4% | 99.2% |
| SAGE (Vals AI) | 51.5% | 28.9% |
| SimpleQA Verified | 63% | 49.3% |
| Vending-Bench 2 (Andon Labs) | 7523.84 | 9047.03 |
| WeirdML (Håvard Tveit Ihle) | 84.9% | 67.3% |
| LMArena Agent (LMArena) | 0.0415 | 0.0128 |
| LMArena Elo | 1466 | 1454 |
| LMArena Vision (LMArena) | 1287 | 1264 |
| SimpleBench (SimpleBench) | 69% | 75.9% |
Higher is better. Each score links to where it was published.