Compare / GLM 4.5 vs GLM 4.6
GLM 4.5 vs GLM 4.6
16 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | GLM 4.5 | GLM 4.6 |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,412 | 1,412 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,372 | 1,400 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,444 | 1,439 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,405 | 1,412 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,411 | 1,424 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,382 | 1,405 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 12.8 | 18.5 |
| Artificial Analysisreported | GPQA Diamond% of questions | 78.2% | 78.0% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 13.0% | 14.5% |
| Artificial Analysisreported | IFBench% of instructions | 44.1% | 43.4% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 52.7% | 54.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 22.0% | 28.8% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 43.0% | 76.9% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 13.0% | 13.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.38 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 57.3 | 53.3 |
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.