Compare / Qwen3 235B A22B Instruct 2507 vs Qwen3 VL 235B A22B Instruct
Qwen3 235B A22B Instruct 2507 vs Qwen3 VL 235B A22B Instruct
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 235B A22B Instruct 2507 | Qwen3 VL 235B A22B Instruct |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,431 | 1,414 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,432 | 1,440 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,378 | 1,357 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,448 | 1,440 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,414 | 1,409 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,422 | 1,413 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,396 | 1,388 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 12.0 | 9.90 |
| Artificial Analysisreported | GPQA Diamond% of questions | 75.3% | 71.2% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 11.1% | 6.6% |
| Artificial Analysisreported | IFBench% of instructions | 46.1% | 42.7% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 33.9% | 32.7% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 15.2% | 6.8% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 33.3% | 35.1% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 15.0% | 19.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.35 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 49.0 | 43.0 |
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.