What I like most is how quickly I can go from a rough idea to something I’d actually use. I mostly create concept art for a side project, and that first batch of four images usually gets me about 70% of the way there. Having consistent style across prompts saves me a lot of rework and makes it easier to keep things cohesive. The pricing feels fair for how much I generate, especially compared with paying for stock images. I also like the Discord and web workflow, since it lets me iterate without much setup. The community gallery ended up being more helpful than I expected, too, especially when I was learning how to phrase prompts. Review collected by and hosted on G2.com.

What I like least is how much the workflow still depends on Discord. Scrolling through a busy channel to find an image I generated last week is tedious, and while the web app has improved things, it still doesn’t fully replace that experience. My biggest frustration is the pricing: once I used up my Fast GPU hours [mid-month / in about X days], the Relax queue made every batch feel slow. Text inside images also still comes out garbled more often than I’d like, which is especially frustrating when I need something like a logo or a label. I'd also like a public API so I can connect it to my own tools. Finally, the first few weeks took longer than they should have because the prompt syntax isn’t explained very well. Review collected by and hosted on G2.com.