domo text to image vs stable diffusion for d&d campaign art

So my D&D group made me basically to be "The Art Guy". As I have just shown you a KI piece and suddenly I am responsible for all graphics in the campaign. I was like Bruh, I didn't want to appear at 2 a.m., so I opened AI tools. First I went with a stable diffusion because it is. I created auto1111, loaded a fantasy model and wrote "Dragonborn Rogue, Candlelit Tavern, Rauch in the background". First render? Disaster. Hands everywhere, the face melted. The second was better, but still not the mood. In the end it did like 7 Gens, optimized CFG, added Loras, Sampler switched. After an hour I finally had something usable. Good art, but I was drained. Then I thought that we could see whether Domo text is easier for the picture. I literally typed "Dragonborn Rogue, who is hiding in the Candelit Tavern." And boom, I had 4 decent pictures in 30 seconds. No settings, no samplers, only vibes. One of them looked so good that I immediately used it in the campaign document. And with Relax Mode Unlimited I became wild. I have generated 15 times and have a whole folder of Tavernenzenen. Some looked roughly grainy, some colorful, but all good enough to throw in our discord. I didn't have to ration loans or stress about "Oh, should I waste this generation". For comparison I also tested Midjourney and do it, why not. MJ gave me beautiful, fantastic things and looked like paintings that they framed on Pinterest boards. The problem is, they were too pretty. My dragonborn looked like a model in a photo shoot that did not hide in a bar. Cool atmosphere, but did not match D&D. So yes: stable diffusion = powerful if you want to find out every slider and want to get in a fine. MJ = aesthetic overload. Domo = fast, practical, fun. Does anyone else use Domo for campaign art? Curious if you combine it for variety.