Compare / qwen3-max-preview vs GPT-5.5
qwen3-max-preview vs GPT-5.5
14 results from 2 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | qwen3-max-preview | GPT-5.5 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,442 | 1,486 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,445 | 1,489 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,393 | 1,452 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,464 | 1,513 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,424 | 1,478 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,435 | 1,481 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,405 | 1,470 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 12.6 | 38.4 |
| Artificial Analysisreported | GPQA Diamond% of questions | 76.4% | 93.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 10.1% | 45.8% |
| Artificial Analysisreported | IFBench% of instructions | 48.0% | 75.8% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 43.0% | 84.3% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 19.7% | 60.6% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 32.7% | 93.9% |
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.