Compare / Claude Sonnet 4.5 vs GPT-5.1
Claude Sonnet 4.5 vs GPT-5.1
14 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 4.5 | GPT-5.1 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,449 | 1,416 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,175 | 1,183 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,449 | 1,459 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,461 | 1,448 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,454 | 1,428 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,498 | 1,482 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,464 | 1,450 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,457 | 1,456 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,454 | 1,437 |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 5.0% | 113.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.41 | 0.38 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 66.9 | 55.8 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 95.6% | 100.0% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 12.0% | 10.9% |
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.