Compare / Qwen3.5 397B A17B vs GPT-5.2
Qwen3.5 397B A17B vs GPT-5.2
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 | GPT-5.2 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,445 | 1,442 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,445 | 1,444 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,405 | 1,391 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,476 | 1,475 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,434 | 1,427 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,442 | 1,437 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,421 | 1,411 |
| tau2-benchreported | Consistency (pass^4)% of tasks | 68.0% | 72.0% |
| tau2-benchreported | Task success (pass^1)% of tasks | 81.5% | 83.0% |
| tau2-benchreported | Consistency (pass^4)% of tasks | 59.6% | 51.8% |
| tau2-benchreported | Task success (pass^1)% of tasks | 84.4% | 81.6% |
| tau2-benchreported | Consistency (pass^4)% of tasks | 92.1% | 71.9% |
| tau2-benchreported | Task success (pass^1)% of tasks | 97.8% | 89.7% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 23.0% | 76.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.33 | 0.40 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 50.1 | 49.4 |
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.