Compare / DeepSeek V3.2 vs GPT-5.2
DeepSeek V3.2 vs GPT-5.2
18 results from 4 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | DeepSeek V3.2 | GPT-5.2 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,437 | 1,442 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,423 | 1,444 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,400 | 1,391 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,453 | 1,475 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,420 | 1,427 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,425 | 1,437 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,410 | 1,411 |
| OpenHands Indexreported | Average% resolved | 45.7% | 58.8% |
| OpenHands Indexreported | Front end (SWE-Bench Multimodal)% resolved | 27.9% | 30.9% |
| OpenHands Indexreported | Greenfield (Commit0)% resolved | 25.0% | 50.0% |
| OpenHands Indexreported | Information gathering (GAIA)% resolved | 50.3% | 65.5% |
| OpenHands Indexreported | Issue resolution (SWE-Bench)% resolved | 71.6% | 74.6% |
| OpenHands Indexreported | Testing (SWT-Bench)% resolved | 53.6% | 73.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 26.0% | 76.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.40 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 57.3 | 49.4 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 92.6% | 100.0% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 6.3% | 8.4% |
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.