Compare / GPT-5.4 vs GLM 4.6
GPT-5.4 vs GLM 4.6
12 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | GPT-5.4 | GLM 4.6 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,485 | 1,447 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,482 | 1,412 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,443 | 1,400 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,515 | 1,439 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,471 | 1,412 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,475 | 1,424 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,465 | 1,405 |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 30.0% | 13.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.40 | 0.38 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 68.8 | 53.3 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 99.9% | 94.5% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 7.0% | 9.5% |
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.