Compare / GLM 4.6 vs GLM 5.2

GLM 4.6 vs GLM 5.2

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

GLM 4.6
Z.ai · profile
GLM 5.2
Z.ai · profile
Results compared
20
Shared sources
3
BenchmarkMetricGLM 4.6GLM 5.2
Arena (formerly LMArena)reportedOverallArena rating 1,4471,463
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4121,463
Arena (formerly LMArena)reportedCreative writingArena rating 1,4001,455
Arena (formerly LMArena)reportedExpert promptsArena rating 1,4391,500
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4121,470
Arena (formerly LMArena)reportedOverallArena rating 1,4241,476
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4051,459
Artificial AnalysisreportedArtificial Analysis Coding Indexindex score 45.868.8
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 18.533.7
Artificial AnalysisreportedGPQA Diamond% of questions 78.0%89.5%
Artificial AnalysisreportedHumanity's Last Exam% of questions 14.5%41.1%
Artificial AnalysisreportedIFBench% of instructions 43.4%73.3%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 54.0%78.3%
Artificial AnalysisreportedTerminal-Bench 2.1% of tasks 49.4%77.9%
Artificial AnalysisreportedTerminal-Bench Hard% of tasks 28.8%50.8%
Artificial AnalysisreportedΤ-bench banking% of tasks 13.4%34.6%
Artificial AnalysisreportedΤ²-bench telecom% of tasks 76.9%99.1%
UGI LeaderboardreportedRequested-length error% off the requested word count 13.0%12.0%
UGI LeaderboardreportedStyle adherencescore from 0 to 1 0.380.37
UGI LeaderboardreportedWriting scorescore out of 100 53.367.0

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.