Compare / Qwen3 VL 235B A22B Instruct vs gpt-oss-120b
Qwen3 VL 235B A22B Instruct vs gpt-oss-120b
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-oss-120b |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,414 | 1,385 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,440 | 1,351 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,357 | 1,277 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,440 | 1,357 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,409 | 1,323 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,413 | 1,352 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,388 | 1,309 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 9.90 | 11.6 |
| Artificial Analysisreported | GPQA Diamond% of questions | 71.2% | 78.2% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 6.6% | 19.6% |
| Artificial Analysisreported | IFBench% of instructions | 42.7% | 69.0% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 32.7% | 52.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 6.8% | 23.5% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 35.1% | 65.8% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 19.0% | 107.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.35 | 0.29 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 43.0 | 38.5 |
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.