GPT-6.1 Sol scores 69.4 on the Respan Index (#4), ahead of Claude Fable 5 at 62.2 (#9). GPT-6.1 Sol leads on 15 of the 24 benchmarks both report. GPT-6.1 Sol is 5.0x cheaper per token at list price. GPT-6.1 Sol has the larger context window (1.1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| FrontierMath Tiers 1-3 v2 (Epoch AI) |
| 87% |
| 93.7% |
| Furniture Assembly (Epoch AI) | 35.8% | 80% |
| GPQA Diamond | 85.9% | 95.4% |
| LiveBench Agentic Coding (LiveBench) | 62.2% | 56.8% |
| LiveBench Coding (LiveBench) | 86% | 80.7% |
| LiveBench Data Analysis (LiveBench) | 80.5% | 82.7% |
| LiveBench Instruction Following (LiveBench) | 75.8% | 74.2% |
| LiveBench Language (LiveBench) | 90.7% | 90.1% |
| LiveBench Mathematics (LiveBench) | 96% | 96.8% |
| LiveBench Reasoning (LiveBench) | 89.7% | 92.6% |
| Mystery Game Puzzles (Epoch AI) | 52% | 80% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 100% |
| SAGE (Vals AI) | 51.9% | 46.5% |
| SimpleQA Verified | 70.7% | 73.9% |
| Terminal-Bench 4.0 (Vals AI) | 41.4% | 55.1% |
| LMArena Agent (LMArena) | 0.0821 | 0.1123 |
| LMArena Elo | 1504 | 1483 |
| LMArena Vision (LMArena) | 1309 | 1291 |
| LMArena WebDev (LMArena) | 1625 | 1758 |
| SimpleBench (SimpleBench) | 81.9% | 82.9% |
Higher is better. Each score links to where it was published.