Compare / GLM 5.1 vs GLM 5.3
GLM 5.1 vs GLM 5.3
19 results from 5 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | GLM 5.1 | GLM 5.3 |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 40.9% | 56.6% |
| Arena (formerly LMArena)reported | OverallArena rating | 1,456 | 1,468 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,465 | 1,474 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,448 | 1,456 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,495 | 1,516 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,460 | 1,480 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,465 | 1,480 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,453 | 1,470 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 55.8 | 74.8 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 26.1 | 44.8 |
| Artificial Analysisreported | GPQA Diamond% of questions | 86.8% | 91.7% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 30.1% | 42.3% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 73.7% | 79.7% |
| Artificial Analysisreported | SciCode% of problems | 44.8% | 59.0% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 61.8% | 83.9% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 2.0% | 41.9% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 13.6% | 50.3% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 34.0% | 41.0% |
| Vending-Bench 2reported | Money after a yearUS dollars | $5,634.41 | $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.