Compare / Qwen3.5-27B vs Qwen3.5 397B A17B
Qwen3.5-27B vs Qwen3.5 397B A17B
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 | Qwen3.5-27B | Qwen3.5 397B A17B |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,420 | 1,445 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,408 | 1,445 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,358 | 1,405 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,441 | 1,476 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,400 | 1,434 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,409 | 1,442 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,380 | 1,421 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 22.9 | 21.4 |
| Artificial Analysisreported | GPQA Diamond% of questions | 85.8% | 89.3% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 23.9% | 29.0% |
| Artificial Analysisreported | IFBench% of instructions | 75.6% | 78.8% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 77.7% | 77.3% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 32.6% | 40.9% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 93.9% | 95.6% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 11.0% | 23.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.33 | 0.33 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 42.4 | 50.1 |
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.