Compare / Claude Sonnet 4 vs DeepSeek V4 Pro 0423
Claude Sonnet 4 vs DeepSeek V4 Pro 0423
11 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Sonnet 4 | DeepSeek V4 Pro 0423 |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,387 | 1,460 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,395 | 1,446 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,433 | 1,480 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,415 | 1,452 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,402 | 1,458 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,398 | 1,451 |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 4.0% | 34.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.36 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 63.7 | 68.4 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 98.6% | 97.2% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 10.3% | 8.6% |
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.