Compare / Qwen3 Coder 480B A35B vs DeepSeek V4 Pro 0423
Qwen3 Coder 480B A35B vs DeepSeek V4 Pro 0423
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 Coder 480B A35B | DeepSeek V4 Pro 0423 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,401 | 1,455 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,391 | 1,460 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,362 | 1,446 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,379 | 1,480 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,384 | 1,452 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,387 | 1,458 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,366 | 1,451 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 11.9 | 36.0 |
| Artificial Analysisreported | GPQA Diamond% of questions | 61.8% | 92.8% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 4.5% | 41.0% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 45.7% | 80.3% |
| OpenHands Indexreported | Average% resolved | 30.9% | 40.7% |
| OpenHands Indexreported | Front end (SWE-Bench Multimodal)% resolved | 23.5% | 36.8% |
| OpenHands Indexreported | Greenfield (Commit0)% resolved | 0.0% | 12.5% |
| OpenHands Indexreported | Information gathering (GAIA)% resolved | 33.9% | 12.7% |
| OpenHands Indexreported | Issue resolution (SWE-Bench)% resolved | 62.4% | 73.2% |
| OpenHands Indexreported | Testing (SWT-Bench)% resolved | 34.9% | 68.1% |
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.