Compare / Claude Fable 5.1 vs GPT-5.2

Claude Fable 5.1 vs GPT-5.2

19 results from 6 sources that measured both models. Each row is in the source's own units; there is no overall winner.

Claude Fable 5.1
Anthropic · profile
GPT-5.2
OpenAI · profile
Results compared
19
Shared sources
6
BenchmarkMetricClaude Fable 5.1GPT-5.2
Arena (formerly LMArena)reportedOverallArena rating 1,4931,425
Arena (formerly LMArena)reportedOverallArena rating 1,5011,442
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4851,444
Arena (formerly LMArena)reportedCreative writingArena rating 1,4841,391
Arena (formerly LMArena)reportedExpert promptsArena rating 1,5161,475
Arena (formerly LMArena)reportedInstruction followingArena rating 1,4971,427
Arena (formerly LMArena)reportedOverallArena rating 1,5011,437
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,4961,411
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 53.430.4
Artificial AnalysisreportedGPQA Diamond% of questions 93.7%90.3%
Artificial AnalysisreportedHumanity's Last Exam% of questions 59.1%37.7%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 85.3%82.7%
Artificial AnalysisreportedΤ-bench banking% of tasks 47.2%11.1%
Remote Labor IndexreportedProjects done to client standard% of projects 17.9%2.5%
SimpleQA Verified (Epoch AI)reportedCorrect answers% of questions 70.8%37.1%
UGI LeaderboardreportedRequested-length error% off the requested word count 3.0%76.0%
UGI LeaderboardreportedStyle adherencescore from 0 to 1 0.430.40
UGI LeaderboardreportedWriting scorescore out of 100 63.949.4
Vending-Bench 2reportedMoney after a yearUS dollars $5,421.56$3,591.33

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.