gpt-6THREAD
测评称 Sol 与 Luna 幻觉率低于前代
Artificial Analysis 称在 max effort 下，GPT-6 Sol 和 Luna 的幻觉率都低于各自前代模型。
Artificial Analysis
@ArtificialAnlys2026.09.23收录原推发布于 2026.09.23
THREAD
GPT-6 Sol and Luna both hallucinate less than their predecessors at max effort.
GPT-6 Sol (max) gains 2 points in the Artificial Analysis Coding Agent Index at half the Cost per Task of its predecessor, while GPT-6 Luna (max) regresses by 2 points.
Artificial Analysis @ArtificialAnlys原推发布于 2026.09.23
x.com/artificialanlys/status/2102462968084758810采集于 2026.09.23 10:21