GPT-5.6 Sol scores 60.6 on the Respan Index (#10), ahead of GPT-5.5 at 52.5 (#15). GPT-5.6 Sol leads on 26 of the 29 benchmarks both report. GPT-5.6 Sol is 1.4x cheaper per token at list price. GPT-5.6 Sol has the larger context window (1.1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| 72.5% |
| 82.9% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 85.3% | 89.1% |
| Furniture Assembly (Epoch AI) | 44.2% | 56.7% |
| GPQA Diamond | 90.7% | 94.6% |
| LiveBench Agentic Coding (LiveBench) | 54% | 56.2% |
| LiveBench Coding (LiveBench) | 82.1% | 83.9% |
| LiveBench Data Analysis (LiveBench) | 81.6% | 79.8% |
| LiveBench Instruction Following (LiveBench) | 70.7% | 71.8% |
| LiveBench Language (LiveBench) | 87.4% | 87.7% |
| LiveBench Mathematics (LiveBench) | 95.9% | 96.2% |
| LiveBench Reasoning (LiveBench) | 89.7% | 91.7% |
| MirrorCode (Epoch AI) | 10% | 20% |
| Mystery Game Puzzles (Epoch AI) | 56% | 58% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 84.4% | 100% |
| SAGE (Vals AI) | 51.5% | 52.6% |
| SimpleQA Verified | 63% | 69.7% |
| Vending-Bench 2 (Andon Labs) | 7523.84 | 9619.37 |
| WeirdML (Håvard Tveit Ihle) | 84.9% | 89.4% |
| LMArena Agent (LMArena) | 0.0415 | 0.0645 |
| LMArena Elo | 1466 | 1484 |
| LMArena Vision (LMArena) | 1287 | 1285 |
| tau2-bench Banking Knowledge (Sierra) | 44.6% | 46.9% |
| GSO Opt@1 (GSO) | 37.3% | 70.6% |
| SimpleBench (SimpleBench) | 69% | 71.7% |
| SWE-Bench Pro | 58.6% | 64.6% |
Higher is better. Each score links to where it was published.