Compare / DeepSeek V4 Pro 0423 vs Grok 4.3
DeepSeek V4 Pro 0423 vs Grok 4.3
21 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 | Grok 4.3 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,455 | 1,428 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,460 | 1,440 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,446 | 1,423 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,480 | 1,452 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,452 | 1,415 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,458 | 1,442 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,451 | 1,424 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 68.8 | 42.2 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 36.0 | 24.9 |
| Artificial Analysisreported | GPQA Diamond% of questions | 92.8% | 90.1% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 41.0% | 37.2% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 80.3% | 75.0% |
| Artificial Analysisreported | SciCode% of problems | 51.0% | 48.3% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 78.7% | 39.7% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 14.1% | 0.0% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 39.6% | 12.4% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 47.0% | 33.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 34.0% | 21.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.34 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 68.4 | 57.7 |
| Vending-Bench 2reported | Money after a yearUS dollars | $3,284.52 | $35.26 |
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.