Compare / Qwen3.5 397B A17B vs gemini-3-pro

Qwen3.5 397B A17B vs gemini-3-pro

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

Qwen3.5 397B A17B
Alibaba · profile
gemini-3-pro
Google · profile
Results compared
20
Shared sources
3
BenchmarkMetricQwen3.5 397B A17Bgemini-3-pro
Arena (formerly LMArena)reportedOverallArena rating 1,4451,481
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4451,474
Arena (formerly LMArena)reportedCreative writingArena rating 1,4051,484
Arena (formerly LMArena)reportedExpert promptsArena rating 1,4761,498
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4341,473
Arena (formerly LMArena)reportedOverallArena rating 1,4421,485
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4211,481
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 21.428.0
Artificial AnalysisreportedGPQA Diamond% of questions 89.3%90.8%
Artificial AnalysisreportedHumanity's Last Exam% of questions 29.0%39.7%
Artificial AnalysisreportedIFBench% of instructions 78.8%70.4%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 77.3%76.0%
Artificial AnalysisreportedTerminal-Bench Hard% of tasks 40.9%41.7%
Artificial AnalysisreportedΤ²-bench telecom% of tasks 95.6%87.1%
tau2-benchreportedConsistency (pass^4)% of tasks 68.0%66.0%
tau2-benchreportedTask success (pass^1)% of tasks 81.5%80.5%
tau2-benchreportedConsistency (pass^4)% of tasks 59.6%47.4%
tau2-benchreportedTask success (pass^1)% of tasks 84.4%75.9%
tau2-benchreportedConsistency (pass^4)% of tasks 92.1%74.6%
tau2-benchreportedTask success (pass^1)% of tasks 97.8%91.0%

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.