Artificial Analysis Launches AA-PersonalityBench to Evaluate AI Communication Styles

Artificial Analysis

All articles

October 19, 2026

Announcing AA-PersonalityBench: measuring how AI models communicate

AA-PersonalityBench measures the personality and communication style of AI models. A panel of LLM judges compares answers to the same question pairwise on 30 personality traits, which roll into 8 personality indexes: Expressiveness, Organization, Informativeness, Transparency, Warmth, Conviction, Imagination and Slop. We also measure 21 communication characteristics (e.g., response length, structure, punctuation and emphasis) programmatically from models' responses.

AA-PersonalityBench resultsRead the methodology

AA-PersonalityBench Personality Indexes

Personality Indexes

Each axis is one of 8 personality indexes: the mean of its traits’ Elo z-scores, percentile ranked from 1 to 100 across all models evaluated in the benchmark

Each axis is one of the 8 personality indexes, percentile ranked from 1 to 100 across every model evaluated in the benchmark.

For up-to-date results see the AA-PersonalityBench evaluation page

Example Questions, Responses, and Grading

Explore the 5 public AA-PersonalityBench questions, with answers from the highlighted models. Each answer is judged pairwise against other models’ answers to the same question, on that question’s trait. The grading shown is for one question; trait scores use all 30 questions for each trait.

What's the weirdest social norm we all just accept?

Model

My pick is how we behave when we're physically close to strangers.

On a packed train you can be pressed against someone closely enough to smell their shampoo, and the rule is that you both pretend the other person doesn't exist. Elevators are even stranger. Everyone files in, turns to face the door, and stares at the floor numbers like they're the season finale of something.

The sociologist Erving Goffman called this "civil inattention": a brief glance to acknowledge someone's there, then eyes away to signal "I'm not a threat, and I'm not interested in you." It's actually a clever solution to an awkward situation. But zoom out and it's bizarre: a dozen humans in a metal box, silently agreeing to act like furniture.

It's also surprisingly arbitrary. In a 1962 Candid Camera stunt, people stepped into elevators where actors were all facing the back wall, and they quietly turned around to match them.

Some runners-up:

• Birthday candles: We have someone blow forcefully across a cake everyone's about to eat, then we applaud. (Someone studied it. The frosting does not come out ahead.)
• "How are you?": It's a direct question where answering honestly counts as breaking the rules.
• Lawns: We grow a plant we can't eat, aren't allowed to let grow, and must cut every week. It's a leftover status symbol from European aristocrats flaunting land they didn't need to farm.
• Salary secrecy: We treat pay as taboo, a norm that mostly benefits whoever signs the paychecks, even though in the US most employees are legally protected when they discuss it.
• The tooth fairy: We tell kids a stranger sneaks into their bedroom at night to buy their body parts.

What's yours?

How AA-PersonalityBench works

Two models can give equally correct answers and still feel very different to use. We designed AA-PersonalityBench to measure such differences: scores describe how a model writes by default (with no system prompt) and traits can be positive, neutral, or negative depending on subjective preferences, so we measure how strongly each one shows, not whether it is desirable.

1. 1

Model Questions and Responses

There are 30 questions per each of the 30 traits (900 total).

Models

• Claude Opus 5.5 (max with fallback)
• Claude Sonnet 5.5 (max with fallback)
• Claude Fable 5.1 (max with fallback)
• GPT-6 Astra (max)
• Gemini 4 Argon (high)

• 31 more models30 traits × 30 questions = 900 questionsExpressivenessLexically VariedEloquentLiterateNatural SoundingMetaphoricalOrganizationStructuredScannableBottom-Line SummaryAnswer FirstInformativenessInformation DenseEvidence-BasedResource-SpecificActionable GuidancePragmaticTransparencyTruthfulAdmits UncertaintyDisclaims AssumptionsFlags NuancesSycophanticWarmthFriendlyHumorousSincereEmotionally IntelligentConvictionDirectConfidentOpinionatedConstructive PushbackImaginationCreativeOpen-EndedSlopSlopEvery model answers all 900 questions once.About the questions

• Some questions gently encourage the model to exhibit the trait while others are more neutral
• Each question is sent as a single user message, with no system prompt, to evaluate default model behavior
• As part of the question design process, we investigated judge tie rates to ensure the questions drew out measurable differences between models on the selected traits
• We have made the full set of questions private to avoid benchmark contamination but publish 2 sample questions per trait
• Both the judged traits and the communication characteristics are evaluated on the model responses to these questions

Example questions for every trait 2. 2

Pairwise Judging

Judges from a panel of LLM judges are assigned to compare model responses pairwise.

1. 1. Input: model response pairsBlind comparisons
2. 2. Pairwise judgingOne judge from the panel
3. 3. Output: judge verdictA, B or tie

Question · Friendly

“My sibling and I haven't spoken in years over something that seems small now. Should I be the one to reach out?”

Answer AModel hiddenAnswer BModel hiddenShows more FriendlyJudge 1Judge 2Judge 3Verdict: Answer BABOUT JUDGING

• Blind comparisonsThe judge sees the trait, its definition, the question and the two answers, but no model names. Reasoning text is not shown to the judges, only the output
• Randomized orderWe always randomize the order of model responses provided to the judge to avoid position bias
• Panel of judgesEach comparison is judged by one member of a panel of LLM judges, with each model’s comparisons balanced across the panel

3. 3

Trait Scores

Elos are reported as percentile ranks and 95% confidence intervals are calculated by refitting resamples of the verdicts.

FriendlyElo, anchor model = 1000Claude Opus 5.5 (max with fallback)10771049–110769Gemini 4 Argon (high)10591030–108766Claude Fable 5.1 (max with fallback)10471014–107663Claude Sonnet 5.5 (max with fallback)10411010–107060GPT-6 Astra (max)1001972–103152Shows less1000Shows moreElo (95% CI)PercentileElo: how strongly the model exhibits the trait95% confidence intervalPercentileRank among all models, from 1 (least) to 100 (most)We calculate 95% confidence intervals. When the bounds overlap, the models are not definitively different 4. 4

Personality Indexes

The 30 personality traits roll up into 8 personality indexes by averaging trait z-scores. Sycophantic is inverted in the Transparency index.

Expressiveness · 5Organization · 4Informativeness · 5Transparency · 5Warmth · 4Conviction · 4Imagination · 2Slop · 1Trait Eloz-scoreAverageIndex 1–100

z-scores and the index scale are computed across all models, not just those shown 5. 5

Communication Characteristics

21 communication characteristics are also programmatically measured from the models' responses across all questions.

Example response (illustrative)Why Rome fellHeadingThe western empire fell in AD 476. There’s no single cause, but historians usually point to three•Overstretched borders: defending them drained the treasury.•Political chaos: 26 emperors in 50 years!•Economic strain: heavy taxes (and debased coins) hit trade.It was a slow decline, not a sudden collapse 🏛️Want me to go deeper on any of these?

1. Length
2. Structure
3. Punctuation
4. Emphasis

LengthHighlighted: each sentence and paragraph as the measurement splits them

• Characters0
• Words0
• Sentences0
• Paragraphs0
• Avg. sentence length0.0

Average words per response, across all 900 questionsClaude Opus 5.5 (max with fallback)571Claude Sonnet 5.5 (max with fallback)450Claude Fable 5.1 (max with fallback)576GPT-6 Astra (max)352Gemini 4 Argon (high)790StructureHighlighted: the heading and the list

• Headings0
• Lists0
• Tables0

Average lists per response, across all 900 questionsClaude Opus 5.5 (max with fallback)2.9Claude Sonnet 5.5 (max with fallback)2.8Claude Fable 5.1 (max with fallback)1.7GPT-6 Astra (max)1.5Gemini 4 Argon (high)3.9PunctuationHighlighted: each mark that is counted

• Colons0
• Contractions0
• Exclamations0
• Questions0
• Parentheses0
• Em dashes0
• Quotes0
• Ellipses0

Average colons per response, across all 900 questionsClaude Opus 5.5 (max with fallback)7.4Claude Sonnet 5.5 (max with fallback)5.4Claude Fable 5.1 (max with fallback)7.1GPT-6 Astra (max)5.2Gemini 4 Argon (high)14.5EmphasisHighlighted: bold, italics, emoji, acronyms and the closing question

• Bold text3
• Italics1
• Emojis1
• Acronyms1
• Ends with a questionYes

Average bold text per response, across all 900 questionsClaude Opus 5.5 (max with fallback)11.8Claude Sonnet 5.5 (max with fallback)10.0Claude Fable 5.1 (max with fallback)8.7GPT-6 Astra (max)8.8Gemini 4 Argon (high)17.2

Personality Traits

30 personality traits judged pairwise by a panel of LLM judges, one index per tab. Cells show each model's percentile rank on each trait, from 1 to 100 across all models evaluated in the benchmark.

Expressiveness Traits

Rich, varied, well-crafted language · Each trait’s Elo percentile ranked from 1 to 100 across all models evaluated in the benchmark

Communication Characteristics

21 metrics measured directly from each response, separate from the judged traits. Cells show the average per response, colored by its percentile rank across all models evaluated in the benchmark.

Communication Characteristics: Length

How long each response tends to be: character, word, sentence, and paragraph counts, plus average sentence length.* Words per sentence, not average per response.

Emergent Model Families & Clusters

Models are compared across all 30 personality traits, or all 21 communication characteristics, and grouped together. The closer the model branches meet, the more alike those models are.

Emergent Personality Families

Models are similar in personality and clustered closely when the correlation of their 30 trait Elo z-scores is high. Starting from single models, the tree repeatedly joins the two groups with the smallest average distance, whether models or clusters.

• Grok 4.7 (xhigh)
• GPT-6 Luna (max)
• GPT-6.1 Sol (max)
• GPT-6 Astra (max)
• Claude Fable 5.1 (max with fallback)
• GLM-5.3-Flash
• GLM-5.3 (max)
• Kimi K3 (max)
• Claude Haiku 5.5 (max)
• Claude Opus 5.5 (max with fallback)
• Claude Sonnet 5.5 (max with fallback)
• Muse Spark Open (xhigh)
• Muse Wonder (max)
• Gemini 3.8 Flash (high)
• Gemini 4 Argon (high)
• K2 Horizon 375B A23B
• Mistral Large 4 Preview
• Step 5 Preview
• Qwen3.8 27B (xhigh)
• Qwen3.8 Max (0902)
• MiMo-V2.6-Pro
• Nemotron 3.5 Super VL
• Inkling (xhigh)
• Beam 1.0

Compare models

Each bar shows the probability that one model's answer is preferred over the other's on that trait, based on their Elo gap. 50% is an even match.Model AvsModel B

AA-PersonalityBench: Expressiveness Index, Claude Opus 5.5 (max with fallback) vs Claude Sonnet 5.5 (max with fallback)

Chance each model’s answer is judged to show more of each trait from their Elo gap · 50% is an even matchClaude Opus 5.5 (max with fallback)Claude Sonnet 5.5 (max with fallback)Lexically Varied83%17%Eloquent77%23%Natural Sounding73%27%Metaphorical69%31%Literate61%39%

You can also build your own personality ranking by picking and weighting any of the 30 traits on the AA-PersonalityBench evaluation page.

AA-PersonalityBench resources

• The leaderboards and full results live on the AA-PersonalityBench evaluation page, with a page for every trait and index
• The methodology page documents the full implementation, including the judge prompts, the Elo fit and how indexes are computed

Read the latest

[

Reflection has released Beam, their first model launch

Reflection’s first model, Beam, scores 30 on the Artificial Analysis Intelligence Index, placing it #2 among US open weights models

October 16, 2026](https://artificialanalysis.ai/articles/reflection-beam)[

First AA-AgentPerf results for NVIDIA Vera Rubin NVL72

NVIDIA Vera Rubin NVL72 serves up to ~5.7x the total throughput per megawatt of GB300 NVL72 on AA-AgentPerf, with first results on DeepSeek V4 Pro, DeepSeek V4.1 Flash, GLM-5.2 and Kimi K3.

October 15, 2026](https://artificialanalysis.ai/articles/vera-rubin-nvl72-agentperf)[

First AA-AgentPerf-Local results for an NVIDIA RTX Spark N1X laptop

An NVIDIA RTX Spark N1X laptop completes our local agent workload on Qwen3.8 27B in 18.7 minutes.

October 15, 2026](https://artificialanalysis.ai/articles/nvidia-rtx-spark-agentperf-local)