AA-PersonalityBench: Gemini models rank most confident and least likely to admit uncertainty
ArtificialAnlys · x · 2026-10-22
Artificial Analysis released AA-PersonalityBench results: Gemini models top the Confident trait — Gemini 3.8 Flash (100), Gemini 4 Argon (97), Gemini 3.1 Pro Preview (95) — and the same three models rank lowest on Admits Uncertainty, scoring just 1, 4, and 6. Inking (high) is the most willing to admit uncertainty at 25. Model personality differences like this have real implications for user trust and perceived hallucination.
More from Models
- Flint MoE model lands in llama.cpp: 39.8B total params, only ~2.5B active per token — jacek2023 · 2026-10-22
- $20 Claude plan with Opus 5.5 feels more generous than the $100 ChatGPT plan, says Christian Selig — rudrank · 2026-10-22
- DeepSeek V4.1 Flash hits 1,521 tok/s on Fireworks Ultrafast, ships Tetris in 11 seconds — nicolechirps · 2026-10-22
- Qwen3.6-35B on a 16GB Mac beats GPT-5-mini on honest tool-calling, at 20x the latency — Mission-Sir-1614 · 2026-10-22
- Ingot quantizes Qwen 27B to FP8 in one week, beating Alibaba's own quantization — const_reborn · 2026-10-22
- Updated code cross-entropy chart: Mirendil-Base edges out Kimi-K3, Beam and Muse — bigblueboo · 2026-10-22