gpt-6POST
GPT-6 Astra High 在 PPTBench 拿 77.34 分
PPTBench 评测显示 Astra High 得分 77.34，但约五分之一任务仍卡在语义关卡，说明视觉编程智能体还有明显短板。
MR. SHAHBAZ
@Mr_Shahbaz_Ai2026.09.30收录原推发布于 2026.09.29
MR. SHAHBAZ @Mr_Shahbaz_Ai原推发布于 2026.09.29
GPT-6 Astra High leads the reported PPTBench score at 77.34, but around one in five tasks still fails a gate. PPTBench breaks those failures down clearly, starting with semantics before checking render/text quality and details. That level of evaluation gives a much clearer picture of what visual coding agents still need to improve.
这条是引用着另一条写的：x.com/einsiaai/status/2104941770433814889
x.com/mr_shahbaz_ai/status/2104953317747282111采集于 2026.09.30 10:24