Compare / GPT-5.4 Mini vs GLM 5.3
GPT-5.4 Mini 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.
| Benchmark | Metric | GPT-5.4 Mini | GLM 5.3 |
|---|---|---|---|
| BulletBenchmeasured by us | Cost per gameUS dollars | $0.0205 | $0.0156 |
| BulletBenchmeasured by us | Games lost on time% of games | 50.0% | 83.3% |
| BulletBenchmeasured by us | Invalid moves% of moves | 0.0% | 0.0% |
| BulletBenchmeasured by us | Ladder Eloladder Elo | 197 | 33 |
| BulletBenchmeasured by us | Median move timemilliseconds | 0.9 s | 1.6 s |
| Arena (formerly LMArena)reported | OverallArena rating | 1,447 | 1,468 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,460 | 1,474 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,402 | 1,456 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,480 | 1,516 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,434 | 1,480 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,447 | 1,480 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,424 | 1,470 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 56.1 | 74.8 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 24.1 | 44.8 |
| Artificial Analysisreported | GPQA Diamond% of questions | 87.5% | 91.7% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 28.1% | 42.3% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 77.0% | 79.7% |
| Artificial Analysisreported | SciCode% of problems | 52.1% | 59.0% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 59.2% | 83.9% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 2.0% | 41.9% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 25.6% | 50.3% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 29.4% | 41.0% |
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.