GLM 5.3 — Benchmarks

Benchmark scores for GLM 5.3 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #23 of 123 open modelscomposite 62.2/100 across 9 benchmarks in 4 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
SciCode56.5%#2 / 50#14 / 155
LiveBench Coding79.0#3 / 16#18 / 54

Knowledge

BenchmarkScoreOpen rankAll models
SimpleQA41.0%#4 / 15#37 / 78

Math

BenchmarkScoreOpen rankAll models
AIME 2024/202591.1%#11 / 74#57 / 270
LiveBench Math87.9#6 / 16#31 / 54
ProofBench49.0%#4 / 21#20 / 64

Reasoning

BenchmarkScoreOpen rankAll models
GPQA Diamond90.9%#7 / 83#30 / 291
LiveBench Reasoning85.8#5 / 16#26 / 54
CritPt19.1%#3 / 50#38 / 167

Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.