Compare / Qwen3.5-35B-A3B vs GPT-5 Mini

Qwen3.5-35B-A3B vs GPT-5 Mini

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

Qwen3.5-35B-A3B
Alibaba · profile
GPT-5 Mini
OpenAI · profile
Results compared
19
Shared sources
3
BenchmarkMetricQwen3.5-35B-A3BGPT-5 Mini
Arena (formerly LMArena)reportedOverallArena rating 1,4081,394
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4031,381
Arena (formerly LMArena)reportedCreative writingArena rating 1,3421,324
Arena (formerly LMArena)reportedExpert promptsArena rating 1,4201,400
Arena (formerly LMArena)reportedInstruction followingArena rating 1,3861,373
Arena (formerly LMArena)reportedOverallArena rating 1,3941,390
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,3641,352
Artificial AnalysisreportedArtificial Analysis Coding Indexindex score 37.015.6
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 19.320.6
Artificial AnalysisreportedGPQA Diamond% of questions 84.5%82.8%
Artificial AnalysisreportedHumanity's Last Exam% of questions 21.0%21.5%
Artificial AnalysisreportedIFBench% of instructions 72.5%75.4%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 72.0%72.3%
Artificial AnalysisreportedTerminal-Bench 2.1% of tasks 40.8%3.7%
Artificial AnalysisreportedTerminal-Bench Hard% of tasks 26.5%33.3%
Artificial AnalysisreportedΤ-bench banking% of tasks 4.9%15.5%
Artificial AnalysisreportedΤ²-bench telecom% of tasks 89.2%71.1%
Vectara Hallucination LeaderboardreportedAnswer rate% of documents 99.8%99.9%
Vectara Hallucination LeaderboardreportedHallucination rate% of summaries 10.5%12.9%

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.