Compare / Gemini 3.1 Pro Preview vs gemini-3-flash

Gemini 3.1 Pro Preview vs gemini-3-flash

15 results from 2 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
gemini-3-flash
Google · profile
Results compared
15
Shared sources
2
BenchmarkMetricGemini 3.1 Pro Previewgemini-3-flash
Arena (formerly LMArena)reportedOverallArena rating 1,4591,439
Arena (formerly LMArena)reportedOverallArena rating 1,2081,191
Arena (formerly LMArena)reportedOverallArena rating 1,4721,468
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4771,469
Arena (formerly LMArena)reportedCreative writingArena rating 1,4801,457
Arena (formerly LMArena)reportedExpert promptsArena rating 1,5091,496
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4791,458
Arena (formerly LMArena)reportedOverallArena rating 1,4871,473
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4801,461
OpenHands IndexreportedAverage% resolved 60.6%49.0%
OpenHands IndexreportedFront end (SWE-Bench Multimodal)% resolved 44.1%22.1%
OpenHands IndexreportedGreenfield (Commit0)% resolved 25.0%18.8%
OpenHands IndexreportedInformation gathering (GAIA)% resolved 81.8%58.8%
OpenHands IndexreportedIssue resolution (SWE-Bench)% resolved 76.8%74.6%
OpenHands IndexreportedTesting (SWT-Bench)% resolved 75.1%70.7%

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.