[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/vt/ - Virtual Youtubers

Search:


View post   

>> No.46271309 [View]
File: 546 KB, 4000x1128, xyz_grid-0010-354885862-best quality, masterpiece, solo, _lora_hll3.1-lora_0.9_,_nanashi mumei, hololive, brown hair, brown eyes, ponytail, hair feather.jpg [View same] [iqdb] [saucenao] [google]
46271309

>>46269805
>Line 1
>Put a "bare" 2girls prompt in here
Might be better to be a bit more clear about this - the first line is the overall prompt for the entire image, so it's usually a good idea to put anything that is relevant to the entire image in it. Background, lighting, style, clothes if they're wearing the same thing, etc
>>46269722
One thing I like about euler a is that is an given step count is relatively fast compared to most other samplers, and even changes to the step count can make noticeable changes to the image. So if you get an image that's really close but isn't quite right, sometimes you can fix it by just nudging the step count up or down a little and trying again. I guess inpainting is an option too but I still have trouble with that sometimes.
Picrel is my attempt at an example, though it's very hard to tell from just a single thing like this. Look at her hands I guess?
I don't use it all the time but I still like using it every so often

Navigation
View posts[+24][+48][+96]