Compare / Qwen3.5 397B A17B vs Claude Sonnet 4.5
Qwen3.5 397B A17B vs Claude Sonnet 4.5
16 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 397B A17B | Claude Sonnet 4.5 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,445 | 1,449 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,445 | 1,461 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,405 | 1,454 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,476 | 1,498 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,434 | 1,464 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,442 | 1,457 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,421 | 1,454 |
| tau2-benchreported | Consistency (pass^4)% of tasks | 68.0% | 48.0% |
| tau2-benchreported | Task success (pass^1)% of tasks | 81.5% | 72.0% |
| tau2-benchreported | Consistency (pass^4)% of tasks | 59.6% | 39.5% |
| tau2-benchreported | Task success (pass^1)% of tasks | 84.4% | 72.4% |
| tau2-benchreported | Consistency (pass^4)% of tasks | 92.1% | 64.0% |
| tau2-benchreported | Task success (pass^1)% of tasks | 97.8% | 84.9% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 23.0% | 5.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.33 | 0.41 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 50.1 | 66.9 |
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.