GPT-6 Sol scores 62.8 on the Respan Index (#8), ahead of DeepSeek V4 Pro at 46.9 (#29). GPT-6 Sol leads on 20 of the 21 benchmarks both report. DeepSeek V4 Pro is 2.0x 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
| 54.9% |
| 52.9% |
| LiveBench Coding (LiveBench) | 77.2% | 81.8% |
| LiveBench Data Analysis (LiveBench) | 79.2% | 81.2% |
| LiveBench Instruction Following (LiveBench) | 67.7% | 68.6% |
| LiveBench Language (LiveBench) | 82.1% | 85.3% |
| LiveBench Mathematics (LiveBench) | 95.1% | 96.4% |
| LiveBench Reasoning (LiveBench) | 85.8% | 88.7% |
| Mystery Game Puzzles (Epoch AI) | 43% | 56% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 96.7% | 100% |
| SimpleQA Verified | 47% | 60.7% |
| Terminal-Bench 4.0 (Vals AI) | 14.1% | 44.4% |
| Vending-Bench 2 (Andon Labs) | 3284.52 | 14427.85 |
| LMArena Agent (LMArena) | 0.0119 | 0.0971 |
| LMArena Elo | 1451 | 1457 |
| LMArena WebDev (LMArena) | 1583 | 1689 |
| DeepSWE v1.1 | 62.7% | 68.8% |
| SimpleBench (SimpleBench) | 50.9% | 73.1% |
Higher is better. Each score links to where it was published.