Compare / Qwen3 VL 235B A22B Instruct vs GPT-5.1
Qwen3 VL 235B A22B Instruct vs GPT-5.1
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 VL 235B A22B Instruct | GPT-5.1 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,414 | 1,459 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,440 | 1,448 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,357 | 1,428 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,440 | 1,482 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,409 | 1,450 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,413 | 1,456 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,388 | 1,437 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 9.90 | 24.7 |
| Artificial Analysisreported | GPQA Diamond% of questions | 71.2% | 87.3% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 6.6% | 28.5% |
| Artificial Analysisreported | IFBench% of instructions | 42.7% | 72.9% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 32.7% | 80.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 6.8% | 45.5% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 35.1% | 81.9% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 19.0% | 113.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.35 | 0.38 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 43.0 | 55.8 |
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.