GPT-6 Sol scores 62.8 on the Respan Index (#8), ahead of Kimi K3 at 50.6 (#21). GPT-6 Sol leads on 17 of the 23 benchmarks both report. GPT-6 Sol is 1.5x cheaper per token at list price. GPT-6 Sol has the larger context window (1.1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| GPQA Diamond |
| 94.3% |
| 93.5% |
| LiveBench Agentic Coding (LiveBench) | 52.9% | 62.2% |
| LiveBench Coding (LiveBench) | 81.8% | 81.4% |
| LiveBench Data Analysis (LiveBench) | 81.2% | 78.7% |
| LiveBench Instruction Following (LiveBench) | 68.6% | 71.4% |
| LiveBench Language (LiveBench) | 85.3% | 85.5% |
| LiveBench Mathematics (LiveBench) | 96.4% | 84.4% |
| LiveBench Reasoning (LiveBench) | 88.7% | 90.7% |
| Mystery Game Puzzles (Epoch AI) | 56% | 26% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 97.2% |
| SAGE (Vals AI) | 44.8% | 52.8% |
| SimpleQA Verified | 60.7% | 50.6% |
| Terminal-Bench 4.0 (Vals AI) | 44.4% | 17.2% |
| Vending-Bench 2 (Andon Labs) | 14427.85 | 5165.04 |
| LMArena Agent (LMArena) | 0.0971 | 0.0418 |
| LMArena Elo | 1457 | 1488 |
| LMArena WebDev (LMArena) | 1689 | 1658 |
| SimpleBench (SimpleBench) | 73.1% | 60.7% |
| DeepSWE v1.1 | 68.8% | 67.3% |
Higher is better. Each score links to where it was published.