Compare / Gemini 3.8 Flash vs Mistral Large 3 2512

Gemini 3.8 Flash vs Mistral Large 3 2512

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

Gemini 3.8 Flash
Google · profile
Mistral Large 3 2512
Mistral · profile
Results compared
19
Shared sources
3
BenchmarkMetricGemini 3.8 FlashMistral Large 3 2512
Arena (formerly LMArena)reportedOverallArena rating 1,4771,434
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4821,409
Arena (formerly LMArena)reportedCreative writingArena rating 1,4821,373
Arena (formerly LMArena)reportedExpert promptsArena rating 1,5291,428
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4881,400
Arena (formerly LMArena)reportedOverallArena rating 1,4921,413
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4861,388
Artificial AnalysisreportedArtificial Analysis Coding Indexindex score 76.320.1
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 40.99.30
Artificial AnalysisreportedGPQA Diamond% of questions 95.3%68.0%
Artificial AnalysisreportedHumanity's Last Exam% of questions 47.8%4.2%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 84.0%36.0%
Artificial AnalysisreportedSciCode% of problems 56.6%36.6%
Artificial AnalysisreportedTerminal-Bench 2.1% of tasks 87.6%12.0%
Artificial AnalysisreportedTerminal-Bench 4.0% of tasks 19.7%0.0%
Artificial AnalysisreportedΤ-bench banking% of tasks 45.8%5.8%
UGI LeaderboardreportedRequested-length error% off the requested word count 13.0%15.0%
UGI LeaderboardreportedStyle adherencescore from 0 to 1 0.340.38
UGI LeaderboardreportedWriting scorescore out of 100 78.541.6

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.