Another one I wanted to do for a very long time just always forgot. Never played the game, but always found her hot!
v2: Redone the whole dataset and adjusted training settings.
A quick note: I'm waiting for the base Z-Image model and for something other than comfy to support it. (If I'd had to use comfy I would've quit this a long time ago)
Recommended style tags:
3d animation, cel rendering, stylizedOutfit: top hat, choker, jacket, cleavage, fingerless gloves, loose belt, mismatched thighhighs, fishnet thighhighs, striped thighhighs, high heel boots
I train everything on IL XL 0.1 so it's compatible with any IL model, but the problem with most checkpoint is that they are leaning towards anime too much and cause the chin and eyes to deform regardless how hard I train it. Legends say it should be allright with NoobAI aswell, I dunno', never tried that, it's an anime wasteland I was told...
Description
FAQ
Comments (14)
Borderlands 2 is amazing. I have nothing but fond memories of it and most female characters are pretty hot. Well all but the one that weighs as much as the entire rest of the female cast 😅
my brother used to play it with his GF, i was too into Tarkov (~4500 hours)
@xJollyboyx BL2 came out way befor Tarkov, like probably 5-7years I think. But damn, 4.5k is a lot. I spent that kind of time on games like Warframe, Tera Online or the old GuildWars. Gets me nostalgic to think about the golden era of MMOs where devs aimed for as many people as possible to like their designs instead of aiming for the least amount of people to take offense
@RedPinkRetro i know, but when my brother used to play it i was playing something else
fuck yesssss
Nice description for the images, mister🤡
you what now?
@xJollyboyx Do you think adding generation data to images demonstrating your models is not necessary? U have around 500!!!!!! models where you don't add the description of generation data. Is it joke?
@Droggoc bye
I'm pretty sure Forge Neo added Z-Image support as a heads-up, but we're still waiting on the base model.
Hopefully in the next few days? Was supposed to be around the end of the weekend.
good to hear that, and yeah i am using forge neo
someone could make a fine tune of it with the tags maybe? like pony or IL, i prefer prompting like that instead of wrighting paragrphs of a story to get an image
@xJollyboyx I know some people are already making LoRAs and finetunes off the Turbo model. AI Toolkit apparently has support for it.
I'm not sure how Booru-style tags will work. Can you even replace the CLIP and still have the model work? I'm not sure.
I do know that there's some people coming up with solutions that will expand your prompt for you, so maybe someone can do an extension or something that'll take the booru tag input and translate it to a more proper description that the CLIP understands (since it's using Qwen 3 4B as its text parser).
nice
why don't you share image prompts and remix?
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.