Compare / Muse Spark 1.1 vs GLM 5.3

Muse Spark 1.1 vs GLM 5.3

22 results from 4 sources that measured both models. Each row is in the source's own units; there is no overall winner.

Muse Spark 1.1
Meta · profile
GLM 5.3
Z.ai · profile
Results compared
22
Shared sources
4
BenchmarkMetricMuse Spark 1.1GLM 5.3
Arena (formerly LMArena)reportedConfirmed task successIPS effect estimate -0.020.06
Arena (formerly LMArena)reportedPraise over complaintIPS effect estimate -0.110.06
Arena (formerly LMArena)reportedSteerabilityIPS effect estimate -0.060.03
Arena (formerly LMArena)reportedTool groundingIPS effect estimate 0.000.00
Arena (formerly LMArena)reportedOverallArena rating 1,4631,468
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4931,474
Arena (formerly LMArena)reportedCreative writingArena rating 1,4491,456
Arena (formerly LMArena)reportedExpert promptsArena rating 1,5041,516
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4731,480
Arena (formerly LMArena)reportedOverallArena rating 1,4911,480
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4601,470
Artificial AnalysisreportedArtificial Analysis Coding Indexindex score 71.374.8
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 33.744.8
Artificial AnalysisreportedGPQA Diamond% of questions 89.8%91.7%
Artificial AnalysisreportedHumanity's Last Exam% of questions 46.2%42.3%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 77.7%79.7%
Artificial AnalysisreportedSciCode% of problems 58.8%59.0%
Artificial AnalysisreportedTerminal-Bench 2.1% of tasks 77.9%83.9%
Artificial AnalysisreportedTerminal-Bench 4.0% of tasks 6.1%41.9%
Artificial AnalysisreportedΤ-bench banking% of tasks 31.8%50.3%
SimpleQA Verified (Epoch AI)reportedCorrect answers% of questions 57.8%41.0%
Vending-Bench 2reportedMoney after a yearUS dollars $6,520.48$8,163.61

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.