LCM does work. But sacrifices overall detail and promp coherence.
No special tricks where used to generated the images, no control net, no multi step process with img2img, no ADetailer, nothing just the basic models in SD WEBUI.
Recommendations
This can be used to generate images of 512/768, 768/512, 512/512 and anything in between, in all my testing no duplicate faces were generated.
Best CFG 7 - 9, i would not go any higher unless you wanna pump up the steps too.
Any checkpoint works but you may have to adjust the weight in order to get it to look right, the max usable weight is 1.3.
I recommend Restart/DPM++ 2M for the first pass and Restart/DPM++ 3M SDE for the Hires.fix pass, this will give you the best results.
Don't go above 2.0 times the first pass resolution, is better to pass it to img2img or use some other tricks to obtain more detail.
Extra notes.
The max resolution i tested is 960, it looks good, but expect duplicated subjects and faces.
If you want better face closeups i suggest you to use ControlNet, sd 1.5 is notorious for generating bad face closeups.
Yet another change to the training process.
Also a bigger and better training set.
And finally some post processing to the model itself, via the SuperMerger plugin for SD WEBUI.
1. Dieses Modell dient nur Lernzwecken. Urheber- und Auslegungsrechte liegen beim Originalautor.
2. Bist du der Originalautor eines Modells, kontaktiere uns bitte zur Authentifizierung über unsere offiziellen Kanäle. Wir schützen die Rechte aller Schöpfer. Hier klicken, um es zu verifizieren.
