In Midjourney you get fantastic results just by using their discord and a text prompt.
To get some similar results in Stable Diffusion you need to set it up, download the models, understand how the various moving parts work together, fiddle with the parameters, donwload specific models out of the hundreds (thousands?) available, iterate, iterate, iterate...
Setting up the environment and tooling around in the code is not a burden, it's a nice change of pace from the boring code I have to deal with normally. Likewise, playing around to build intuition about how prompts and parameters correspond to neighborhoods in latent space is quite fun.
Beyond that, being able to go to sleep with my computer doing a massive batch job state space exploration and wake up with a bunch of cool stuff to look at gives me Christmas vibes daily.
Sure, but if Midjourney outputs a low quality results for your prompt, they are going to be much more difficult to improve. It's a black box at this point.
While with SD there can be multiple solutions for a single problem, but yeah, you have to develop your own workflow (which will inevitably break with new updates)
In Midjourney you get fantastic results just by using their discord and a text prompt.
To get some similar results in Stable Diffusion you need to set it up, download the models, understand how the various moving parts work together, fiddle with the parameters, donwload specific models out of the hundreds (thousands?) available, iterate, iterate, iterate...