Compare / Qwen3.5 397B A17B vs Kimi K3
Qwen3.5 397B A17B vs Kimi K3
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 397B A17B | Kimi K3 |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 24.9% | 50.6% |
| Arena (formerly LMArena)reported | OverallArena rating | 1,445 | 1,472 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,445 | 1,489 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,405 | 1,458 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,476 | 1,533 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,434 | 1,486 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,442 | 1,488 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,421 | 1,475 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 48.2 | 76.2 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 21.4 | 43.6 |
| Artificial Analysisreported | GPQA Diamond% of questions | 89.3% | 93.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 29.0% | 46.9% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 77.3% | 88.7% |
| Artificial Analysisreported | SciCode% of problems | 44.8% | 59.5% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 51.3% | 85.0% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 0.0% | 12.6% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 13.4% | 46.0% |
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.