GPT-5.6 Sol scores 60.6 on the Respan Index (#10), ahead of GPT-5.6 Terra at 53.5 (#14). GPT-5.6 Sol leads on 27 of the 28 benchmarks both report. GPT-5.6 Terra is 1.8x cheaper per token at list price.
* Estimated: no published scores in that category. How the Respan Index works
| 89.1% |
| 86% |
| Furniture Assembly (Epoch AI) | 56.7% | 54.2% |
| GPQA Diamond | 94.6% | 92.9% |
| LiveBench Agentic Coding (LiveBench) | 56.2% | 54.9% |
| LiveBench Coding (LiveBench) | 83.9% | 78.2% |
| LiveBench Data Analysis (LiveBench) | 79.8% | 79.3% |
| LiveBench Instruction Following (LiveBench) | 71.8% | 64.6% |
| LiveBench Language (LiveBench) | 87.7% | 82.9% |
| LiveBench Mathematics (LiveBench) | 96.2% | 94.9% |
| LiveBench Reasoning (LiveBench) | 91.7% | 90.6% |
| Mystery Game Puzzles (Epoch AI) | 58% | 35% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 99.7% |
| SAGE (Vals AI) | 52.6% | 47% |
| SimpleQA Verified | 69.7% | 43.2% |
| Terminal-Bench 4.0 (Vals AI) | 37.9% | 22.7% |
| Vending-Bench 2 (Andon Labs) | 9619.37 | 7343.21 |
| WeirdML (Håvard Tveit Ihle) | 89.4% | 78.3% |
| LMArena Agent (LMArena) | 0.0645 | 0.004 |
| LMArena Elo | 1484 | 1466 |
| LMArena Vision (LMArena) | 1285 | 1268 |
| BALROG (BALROG) | 60% | 53.2% |
| DeepSWE v1.1 | 72.7% | 69.6% |
| SimpleBench (SimpleBench) | 71.7% | 48.9% |
| SWE-Bench Pro | 64.6% | 63.4% |
Higher is better. Each score links to where it was published.