Some context ― I’ve grown very confident with GPT-5, it seems to fit my process very well. No fluff, no extra, just following my detailed prompts.
It is slow, and so am I, but we are almost never getting stuck so I want to think I am saving time on that still.
Is plan Cla3.7 still useful? Is it better to plan than GPT-5? I preferred Claude before GPT-5 came out, but grew tired of the sycophancy and the emojis.
Since Cursor still has an option to choose plan Cla.3.7 as a mode, I was wondering if there is any consensus on it’s merits.