Leaderboards

GeminiGemini 4 Argon

google/gemini-4-argon
Overall score
3.390
95% CI [3.361, 3.418]
Based on 4,528 sessions
Rank 3 of 13 models on the Chinese leaderboard
4 aspects
● Conversational ability 0%4.03
● Character consistency 100%3.39
● Role-playing attractiveness 0%3.67
● Personality back-testing 0%—
13 dimensions
Fluency
3.85
Coherency
4.20
Consistency
4.02
Knowledge exposure
2.93
Knowledge accuracy
3.33
Knowledge hallucination
3.42
Persona behavior
3.83
Persona utterance
3.43
Human-likeness
3.79
Communication skills
3.84
Expression diversity
3.48
Empathy
3.56
MBTI accuracy
—
FluencyCoherencyConsistencyKnowledge exposureKnowledge accuracyKnowledge hallucinationPersona behaviorPersona utteranceHuman-likenessCommunication skillsExpression diversityEmpathyMBTI accuracy2

Scores by language

Tested November 1, 2026
LanguageRankOverall95% confidence intervalConversational abilityCharacter consistencyRole-playing attractivenessPersonality back-testingSessions
Chinese3 / 133.393.4 – 3.44.03.43.7—4,528