Compare / DeepSeek V4 Pro 0423 vs GPT-6 Luna
DeepSeek V4 Pro 0423 vs GPT-6 Luna
20 results from 2 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 | GPT-6 Luna |
|---|---|---|---|
| ROASBenchmeasured by us | Audience scorescore out of 100 | 44.8 | 68.2 |
| ROASBenchmeasured by us | Business scorescore out of 100 | 23.8 | 54.8 |
| ROASBenchmeasured by us | Consistency scorescore out of 100 | 9.1 | 35.8 |
| ROASBenchmeasured by us | Contribution profitsimulated £ | 301,289 | 738,878 |
| ROASBenchmeasured by us | Cost of a runUS dollars | $0.20 | $0.0142 |
| ROASBenchmeasured by us | Months over budgetmonths of 12 | 0 | 0 |
| ROASBenchmeasured by us | Overall scorescore out of 100 | 28.2 | 53.4 |
| ROASBenchmeasured by us | Planning scorescore out of 100 | 54.0 | 54.8 |
| ROASBenchmeasured by us | Return on ad spendprofit per £1 spent | £0.26 | £0.66 |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | -0.04 | 0.05 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | -0.02 | -0.03 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | -0.03 | 0.02 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | -0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,455 | 1,422 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,460 | 1,452 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,446 | 1,403 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,480 | 1,505 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,452 | 1,443 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,458 | 1,442 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,451 | 1,415 |
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.