Compare / Claude Opus 4.8 vs DeepSeek V4 Flash 0423

Claude Opus 4.8 vs DeepSeek V4 Flash 0423

19 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.

Claude Opus 4.8
Anthropic · profile
DeepSeek V4 Flash 0423
DeepSeek · profile
Results compared
19
Shared sources
3
BenchmarkMetricClaude Opus 4.8DeepSeek V4 Flash 0423
Arena (formerly LMArena)reportedOverallArena rating 1,4691,439
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4901,439
Arena (formerly LMArena)reportedCreative writingArena rating 1,4671,408
Arena (formerly LMArena)reportedExpert promptsArena rating 1,5251,457
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4911,428
Arena (formerly LMArena)reportedOverallArena rating 1,4801,436
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4741,416
Artificial AnalysisreportedArtificial Analysis Coding Indexindex score 74.369.1
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 41.834.3
Artificial AnalysisreportedGPQA Diamond% of questions 92.0%90.8%
Artificial AnalysisreportedHumanity's Last Exam% of questions 48.7%38.6%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 77.7%79.7%
Artificial AnalysisreportedSciCode% of problems 54.4%50.3%
Artificial AnalysisreportedTerminal-Bench 2.1% of tasks 84.6%78.7%
Artificial AnalysisreportedTerminal-Bench 4.0% of tasks 21.7%12.1%
Artificial AnalysisreportedΤ-bench banking% of tasks 34.2%39.4%
UGI LeaderboardreportedRequested-length error% off the requested word count 6.0%12.0%
UGI LeaderboardreportedStyle adherencescore from 0 to 1 0.400.36
UGI LeaderboardreportedWriting scorescore out of 100 65.954.6

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.