Compare / Gemma 4 26B A4B vs GLM 5.2

Gemma 4 26B A4B vs GLM 5.2

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

Gemma 4 26B A4B
Google · profile
GLM 5.2
Z.ai · profile
Results compared
21
Shared sources
3
BenchmarkMetricGemma 4 26B A4BGLM 5.2
Arena (formerly LMArena)reportedOverallArena rating 1,4451,463
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4441,463
Arena (formerly LMArena)reportedCreative writingArena rating 1,3991,455
Arena (formerly LMArena)reportedExpert promptsArena rating 1,4751,500
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4371,470
Arena (formerly LMArena)reportedOverallArena rating 1,4381,476
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4061,459
Artificial AnalysisreportedArtificial Analysis Coding Indexindex score 39.368.8
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 16.733.7
Artificial AnalysisreportedGPQA Diamond% of questions 79.2%89.5%
Artificial AnalysisreportedHumanity's Last Exam% of questions 19.3%41.1%
Artificial AnalysisreportedIFBench% of instructions 72.4%73.3%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 65.7%78.3%
Artificial AnalysisreportedSciCode% of problems 40.0%51.2%
Artificial AnalysisreportedTerminal-Bench 2.1% of tasks 39.0%77.9%
Artificial AnalysisreportedTerminal-Bench Hard% of tasks 25.0%50.8%
Artificial AnalysisreportedΤ-bench banking% of tasks 12.0%34.6%
Artificial AnalysisreportedΤ²-bench telecom% of tasks 43.6%99.1%
UGI LeaderboardreportedRequested-length error% off the requested word count 8.0%12.0%
UGI LeaderboardreportedStyle adherencescore from 0 to 1 0.350.37
UGI LeaderboardreportedWriting scorescore out of 100 43.867.0

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.