Compare / Mistral Small 3 vs o4 Mini

Mistral Small 3 vs o4 Mini

11 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.

Mistral Small 3
Mistral · profile
o4 Mini
OpenAI · profile
Results compared
11
Shared sources
3
BenchmarkMetricMistral Small 3o4 Mini
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,2591,379
Arena (formerly LMArena)reportedCreative writingArena rating 1,2271,337
Arena (formerly LMArena)reportedExpert promptsArena rating 1,2591,408
Arena (formerly LMArena)reportedInstruction followingArena rating 1,2551,369
Arena (formerly LMArena)reportedOverallArena rating 1,2741,391
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,2481,354
UGI LeaderboardreportedRequested-length error% off the requested word count 8.0%19.0%
UGI LeaderboardreportedStyle adherencescore from 0 to 1 0.350.33
UGI LeaderboardreportedWriting scorescore out of 100 33.946.9
Vectara Hallucination LeaderboardreportedAnswer rate% of documents 97.9%98.7%
Vectara Hallucination LeaderboardreportedHallucination rate% of summaries 5.1%18.6%

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.