GPT-6.1 Sol scores 69.4 on the Respan Index (#4), ahead of Qwen3.8-Max at 50.3 (#22). GPT-6.1 Sol leads on 18 of the 22 benchmarks both report. Qwen3.8-Max is 1.3x 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
| Furniture Assembly (Epoch AI) |
| 80% |
| 20% |
| GPQA Diamond | 95.4% | 92.6% |
| LiveBench Agentic Coding (LiveBench) | 56.8% | 64.6% |
| LiveBench Coding (LiveBench) | 80.7% | 72.9% |
| LiveBench Data Analysis (LiveBench) | 82.7% | 78.4% |
| LiveBench Instruction Following (LiveBench) | 74.2% | 74.1% |
| LiveBench Language (LiveBench) | 90.1% | 79.7% |
| LiveBench Mathematics (LiveBench) | 96.8% | 91.3% |
| LiveBench Reasoning (LiveBench) | 92.6% | 88.2% |
| Mystery Game Puzzles (Epoch AI) | 80% | 38% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 99.4% |
| SAGE (Vals AI) | 46.5% | 51.3% |
| SimpleQA Verified | 73.9% | 45.8% |
| Terminal-Bench 4.0 (Vals AI) | 55.1% | 34.3% |
| LMArena Agent (LMArena) | 0.1123 | 0.0251 |
| LMArena Elo | 1483 | 1481 |
| LMArena Vision (LMArena) | 1291 | 1301 |
| LMArena WebDev (LMArena) | 1758 | 1671 |
Higher is better. Each score links to where it was published.