Compare / GPT-5.2 vs GLM 4.7 Flash
GPT-5.2 vs GLM 4.7 Flash
19 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-5.2 | GLM 4.7 Flash |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,442 | 1,370 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,444 | 1,368 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,391 | 1,306 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,475 | 1,386 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,427 | 1,348 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,437 | 1,365 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,411 | 1,331 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 30.4 | 14.9 |
| Artificial Analysisreported | GPQA Diamond% of questions | 90.3% | 58.1% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 37.7% | 7.6% |
| Artificial Analysisreported | IFBench% of instructions | 75.4% | 60.8% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 82.7% | 41.7% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 47.0% | 22.0% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 84.8% | 98.8% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 76.0% | 39.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.40 | 0.37 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 49.4 | 25.0 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 100.0% | 91.6% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 8.4% | 9.3% |
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.