Compare / Gemini 3.1 Pro Preview vs Mistral Small 4

Gemini 3.1 Pro Preview vs Mistral Small 4

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

Gemini 3.1 Pro Preview
Google · profile
Mistral Small 4
Mistral · profile
Results compared
20
Shared sources
3
BenchmarkMetricGemini 3.1 Pro PreviewMistral Small 4
BulletBenchmeasured by usCost per gameUS dollars $0.0446$0.0039
BulletBenchmeasured by usGames lost on time% of games 58.3%25.0%
BulletBenchmeasured by usInvalid moves% of moves 0.0%0.6%
BulletBenchmeasured by usLadder Eloladder Elo 529367
BulletBenchmeasured by usMedian move timemilliseconds 3.1 s0.6 s
Artificial AnalysisreportedArtificial Analysis Coding Indexindex score 68.826.6
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 29.711.3
Artificial AnalysisreportedGPQA Diamond% of questions 94.1%76.9%
Artificial AnalysisreportedHumanity's Last Exam% of questions 47.0%9.9%
Artificial AnalysisreportedIFBench% of instructions 77.1%48.2%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 82.0%49.7%
Artificial AnalysisreportedSciCode% of problems 58.7%38.8%
Artificial AnalysisreportedTerminal-Bench 2.1% of tasks 73.8%21.0%
Artificial AnalysisreportedTerminal-Bench 4.0% of tasks 4.0%0.0%
Artificial AnalysisreportedTerminal-Bench Hard% of tasks 53.8%17.4%
Artificial AnalysisreportedΤ-bench banking% of tasks 21.4%4.9%
Artificial AnalysisreportedΤ²-bench telecom% of tasks 95.6%41.2%
UGI LeaderboardreportedRequested-length error% off the requested word count 19.0%22.0%
UGI LeaderboardreportedStyle adherencescore from 0 to 1 0.340.31
UGI LeaderboardreportedWriting scorescore out of 100 72.240.3

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.