Compare / Qwen3.7 Max vs GPT-5.6 Sol
Qwen3.7 Max vs GPT-5.6 Sol
17 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Qwen3.7 Max | GPT-5.6 Sol |
|---|---|---|---|
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | -0.09 | 0.03 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | -0.08 | 0.19 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | -0.03 | 0.07 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | -0.00 | 0.00 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 66.0 | 78.3 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 29.5 | 47.0 |
| Artificial Analysisreported | GPQA Diamond% of questions | 92.3% | 94.1% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 40.5% | 49.5% |
| Artificial Analysisreported | IFBench% of instructions | 80.5% | 72.7% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 79.0% | 84.0% |
| Artificial Analysisreported | SciCode% of problems | 49.5% | 57.8% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 74.5% | 89.5% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 1.5% | 39.9% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 50.8% | 65.9% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 11.8% | 44.3% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 94.7% | 85.1% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 55.8% | 69.7% |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.