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
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| SciCode | 56.5% | #2 / 50 | #14 / 155 |
| LiveBench Coding | 79.0 | #3 / 16 | #18 / 54 |
Knowledge
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| SimpleQA | 41.0% | #4 / 15 | #37 / 78 |
Math
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| AIME 2024/2025 | 91.1% | #11 / 74 | #57 / 270 |
| LiveBench Math | 87.9 | #6 / 16 | #31 / 54 |
| ProofBench | 49.0% | #4 / 21 | #20 / 64 |
Reasoning
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| GPQA Diamond | 90.9% | #7 / 83 | #30 / 291 |
| LiveBench Reasoning | 85.8 | #5 / 16 | #26 / 54 |
| CritPt | 19.1% | #3 / 50 | #38 / 167 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.