GLM-5.3 Tops Em Dash Usage Count While Claude Barely Uses Any

Artificial Analysis's new AA-PersonalityBench reveals stylistic differences among models: GLM-5.3 averages 7.4 em dashes per response, the highest of any model, while Claude uses almost none.

2026-11-05 ~ 2026-11-05 · 2 related posts

• GLM-5.3 averages 7.4 em dashes per response; Claude models use almost none — ArtificialAnlys · 2026-11-05
• AA-PersonalityBench: GLM-5.3 averages 7.4 em dashes per reply, Claude almost none — ArtificialAnlys · 2026-11-05