GPT-5.6 Sol scores 60.6 on the Respan Index (#10), ahead of Claude Opus 4.8 at 52.2 (#16). GPT-5.6 Sol leads on 24 of the 29 benchmarks both report. GPT-5.6 Sol is 1.3x 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
| 56.1% |
| 82.9% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 80% | 89.1% |
| Furniture Assembly (Epoch AI) | 42.5% | 56.7% |
| GPQA Diamond | 93.6% | 94.6% |
| LiveBench Agentic Coding (LiveBench) | 50.5% | 56.2% |
| LiveBench Coding (LiveBench) | 81.8% | 83.9% |
| LiveBench Data Analysis (LiveBench) | 66% | 79.8% |
| LiveBench Instruction Following (LiveBench) | 72% | 71.8% |
| LiveBench Language (LiveBench) | 79.7% | 87.7% |
| LiveBench Mathematics (LiveBench) | 94.3% | 96.2% |
| LiveBench Reasoning (LiveBench) | 89.2% | 91.7% |
| Mystery Game Puzzles (Epoch AI) | 36% | 58% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 98.3% | 100% |
| SAGE (Vals AI) | 54.8% | 52.6% |
| SimpleQA Verified | 53% | 69.7% |
| Terminal-Bench 4.0 (Vals AI) | 23.2% | 37.9% |
| Vending-Bench 2 (Andon Labs) | 5787.43 | 9619.37 |
| WeirdML (Håvard Tveit Ihle) | 82.9% | 89.4% |
| LMArena Agent (LMArena) | 0.0664 | 0.0645 |
| LMArena Elo | 1453 | 1484 |
| LMArena Vision (LMArena) | 1286 | 1285 |
| tau2-bench Banking Knowledge (Sierra) | 39.7% | 46.9% |
| GSO Opt@1 (GSO) | 47.1% | 70.6% |
| SimpleBench (SimpleBench) | 64.8% | 71.7% |
| SWE-Bench Pro | 69.2% | 64.6% |
Higher is better. Each score links to where it was published.