Compare / DeepSeek V4 Pro 0423 vs Grok 4.7

DeepSeek V4 Pro 0423 vs Grok 4.7

20 results from 2 sources that measured both models. Each row is in the source's own units; there is no overall winner.

DeepSeek V4 Pro 0423
DeepSeek · profile
Grok 4.7
xAI · profile
Results compared
20
Shared sources
2
BenchmarkMetricDeepSeek V4 Pro 0423Grok 4.7
ROASBenchmeasured by usAudience scorescore out of 100 44.858.2
ROASBenchmeasured by usBusiness scorescore out of 100 23.826.9
ROASBenchmeasured by usConsistency scorescore out of 100 9.130.5
ROASBenchmeasured by usContribution profitsimulated US dollars $301,289$187,636
ROASBenchmeasured by usCost of a runUS dollars $0.20$0.42
ROASBenchmeasured by usMonths over budgetmonths of 12 00
ROASBenchmeasured by usOverall scorescore out of 100 28.236.6
ROASBenchmeasured by usPlanning scorescore out of 100 54.054.6
ROASBenchmeasured by usReturn on ad spendprofit per $1 spent $0.26$0.23
Arena (formerly LMArena)reportedConfirmed task successIPS effect estimate -0.040.09
Arena (formerly LMArena)reportedPraise over complaintIPS effect estimate -0.020.04
Arena (formerly LMArena)reportedSteerabilityIPS effect estimate -0.030.00
Arena (formerly LMArena)reportedTool groundingIPS effect estimate -0.000.00
Arena (formerly LMArena)reportedOverallArena rating 1,4551,420
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4601,424
Arena (formerly LMArena)reportedCreative writingArena rating 1,4461,427
Arena (formerly LMArena)reportedExpert promptsArena rating 1,4801,454
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4521,437
Arena (formerly LMArena)reportedOverallArena rating 1,4581,439
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4511,426

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.