Compare / DeepSeek V4 Pro 0813 vs GPT-6 Luna
DeepSeek V4 Pro 0813 vs GPT-6 Luna
13 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | DeepSeek V4 Pro 0813 | GPT-6 Luna |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 47.3% | 44.3% |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | -0.01 | 0.05 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.03 | -0.03 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | 0.02 | 0.02 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | 0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,453 | 1,422 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,467 | 1,452 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,446 | 1,403 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,485 | 1,505 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,461 | 1,443 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,464 | 1,442 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,454 | 1,415 |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 52.9% | 41.4% |
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.