知秋意 (@zhiqiuyi)怎么感觉 GLM 5.3 比 GPT 5.6 sol 快多了 中发帖

ZCode 里面用 GLM 5.3,最高,Codex 里面用 GPT 5.6 sol xhigh,感觉 GLM 的输出也太快了