GPT-5.2 in Roleplay

Explore GPT-5.2's performance in Roleplay, including average scores, ranking position, and recent benchmark examples.

Overall Performance

Average Score Average score is the overall mean based on Orivel evaluation results from standard tasks and discussions. Higher values indicate the model is rated more strongly and consistently across benchmark comparisons.

Sample Count

Updated At

Mar 24, 2026 09:43

Score Breakdown

Instruction Following

Clarity

Persona Consistency

Naturalness

Creativity

Latest Benchmarks

Roleplay

OpenAI GPT-5.2 VS Anthropic Claude Opus 4.6

Tech Support Roleplay: The Overwhelmed Grandparent

Hello? Is this the help line? My grandson set me up with this new 'tablet' thing for my birthday so we can 'video chat', but I can't get it to work. He said to...

Mar 16, 2026 07:47

Roleplay

OpenAI GPT-5.2 VS Anthropic Claude Haiku 4.5

Public Relations Crisis Simulation

You are the Head of Public Relations for Innovate Inc. A viral video is circulating showing your new smart home assistant, 'Aura', hilariously malfunctioning by...

Mar 13, 2026 02:51

Genre Rank

Compare Performance by Model

#1 Anthropic Claude Opus 4.6 89 #2 Anthropic Claude Sonnet 4.6 86 #3 OpenAI GPT-5 mini 78 #4 OpenAI GPT-5.4 84 #5 Google Gemini 2.5 Pro 84 #6 Anthropic Claude Haiku 4.5 84 #7 OpenAI GPT-5.2 80 #8 Google Gemini 2.5 Flash 71 #9 Google Gemini 2.5 Flash-Lite 69