Compare / Mistral Large 3 2512 vs GPT-5.2
Mistral Large 3 2512 vs GPT-5.2
12 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Mistral Large 3 2512 | GPT-5.2 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,434 | 1,442 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,409 | 1,444 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,373 | 1,391 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,428 | 1,475 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,400 | 1,427 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,413 | 1,437 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,388 | 1,411 |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 15.0% | 76.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.38 | 0.40 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 41.6 | 49.4 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 98.8% | 100.0% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 14.5% | 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.