October 19, 2026

AA-PersonalityBench: models with correct answers still sound different, and now 30 traits are measured

Artificial Analysis released AA-PersonalityBench on Oct 19: 30 personality traits, 21 style markers and 900 questions. One difference already shows in the numbers: Qwen3.8 27B writes about 1,030 words per answer on average, nearly four times more than GPT-6 Luna (270).

Source

@artificialanlys · X · x.com ↗

who it's about

DeepSeek →GPT-6 Astra →Claude Fable 5.1 →Grok 4.7 →GPT-6 Luna →GLM-5.3 →Claude Opus 5.5 →Claude Sonnet 5.5 →OpenAI →Anthropic →xAI →Zhipu →

about the same entities

News
Anthropic nears second $45 billion Nscale deal, Bloomberg reportsOctober 20
News
OpenAI says its model got better at COBOL, Fortran and Visual Basic, but published no benchmarksOctober 20
Release
/auto-mode-setup in Claude Code: setup wizard now works on Bedrock tooOctober 20

All news

One email a week - only what changes how you work. Subscribe

News