Compare / DeepSeek V4 Pro 0423 vs GLM 5.2
DeepSeek V4 Pro 0423 vs GLM 5.2
25 results from 5 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | DeepSeek V4 Pro 0423 | GLM 5.2 |
|---|---|---|---|
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | -0.04 | 0.04 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | -0.02 | 0.10 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | -0.03 | 0.04 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | -0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,455 | 1,463 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,460 | 1,463 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,446 | 1,455 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,480 | 1,500 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,452 | 1,470 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,458 | 1,476 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,451 | 1,459 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 68.8 | 68.8 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 36.0 | 33.7 |
| Artificial Analysisreported | GPQA Diamond% of questions | 92.8% | 89.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 41.0% | 41.1% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 80.3% | 78.3% |
| Artificial Analysisreported | SciCode% of problems | 51.0% | 51.2% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 78.7% | 77.9% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 14.1% | 1.0% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 39.6% | 34.6% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 47.0% | 34.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 34.0% | 12.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.37 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 68.4 | 67.0 |
| Vending-Bench 2reported | Money after a yearUS dollars | $3,284.52 | $8,313.78 |
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.