Artificial Analysis Launches PersonalityBench; GPT-6 Astra Scores 16/100 on Slop, Sparking Debate
max_paperclips · x · 2026-10-22
- Artificial Analysis released AA-PersonalityBench, a benchmark using 900 curated questions to measure 30 personality traits and 21 communication-style characteristics across LLMs, aiming to capture how two equally correct models can still feel very different in use (imagination for creative writing vs. directness for legal work).
- A chart shows GPT-6 Astra (max) scoring just 16/100 on the Slop dimension (95% CI 12-18).
- Critic @lossgobbler blasted AA's definition of "slop," arguing ultra-precise bloviating is actually "kino," and disputed the bench's rankings of MiMo, Grok and Gemini.
More from Models
- Pichai says internal results show Gemini 5 will be a big jump over Argon — rohanpaul_ai · 2026-10-22
- RAG explainer: real-time retrieval fixes stale data and hallucinations without fine-tuning — goyalshaliniuk · 2026-10-22
- Chat template reversal experiment finds Mistral Small 3 24B best at human mimicry — EntropyMagnets · 2026-10-22
- Founder banned from Claude two days after annual renewal despite $1,500 monthly spend — dotey · 2026-10-22
- MiniMax M3.1 Flash tested on 3D whale-skeleton city, claims 70% cost cut — HeyZoyaKhan · 2026-10-22
- OpenAI's Astra tops SWE-sweep bug-hunting leaderboard at less than half the cost of 5.6 — OfirPress · 2026-10-22