Compare / DeepSeek V3.1 vs gpt-oss-120b
DeepSeek V3.1 vs gpt-oss-120b
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 | DeepSeek V3.1 | gpt-oss-120b |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,417 | 1,351 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,402 | 1,277 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,432 | 1,357 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,416 | 1,323 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,417 | 1,352 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,407 | 1,309 |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 30.0% | 107.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.29 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 51.9 | 38.5 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 94.5% | 99.9% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 5.5% | 14.2% |
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.