KREA2 Gomes v3: Slightly different body, from v1, arm and side torso tattoos, same peen as v1 dataset - lighter dark skin, cropped hair, more of a basketball player build, big fat dick, and low hanger balls.

KREA2 Kang v1: Different body and peen dataset - dark skinned, dreds, muscular build, long slender dick, and smooth balls.

KREA2 Gomes v1: Mostly the same dataset as the last but I decided to comb through the captions to make sure that arrosal state was properly captioned - so now when you specify erect or flaccid, you should get the dong you were expecting instead of a roll of the dice. I also cleaned up some other areas of the captioning, mostly pronouns, so the result is a much more gayy character if prompt to uh, interact, with another dude. Fair warning if you do prompt such debauchery - while facial identity won't bleed too much, but the tattoos might - so carefully construct your prompts accordingly.

With this dataset, here are some prompt adjustable characteristics I've tested with suckess :

✅ Custom collar tattoo wording

✅ Erect / flaccid peen (with or without nut)

✅ Buttsecks

✅ Knob slobbin'

✅ Jockstraps (various angles)

🟡 Bling him out w/ a cock ring

Note: The above aren't necessarily trigger words, just prompt naturally and it should work. See gallery for examples.

--Flux 2 Klein versions--

v2.1: This is a lora merge of a v2 and a very similar character which used much higher quality dataset even than v2. This is my first attempt to generate a new AI character using lora merging but the result is that generations are less likely to add tattoos unless specified by prompting giving it a bit more flexibility. Other anatomical factors remain fairly consistent so that's why I think of this more of a variant than a net-new model.

v2: Upscaled and refined dataset from previous version which yields higher HQ generations.

I created an AI Nobody character using Flux 2 Klein to mostly see how some of my previous amalgamated datasets and previous captioning, to see what would result with this newer model. This model was more of a test to see how much fidelity/accuracy I can get with lower bit quantization which the results surprised me how well it learned even at low bit.

I'm still playing around w/ this Flux 2 Klein but thought I'd share this given that it's a fabricated persona and with it being a 4B it can be used without BFL licensing restrictions.

What can it be used for? Well aside from the obvious full character consistency purposes, which let's face it - there are plenty of those at this point and it's easy enough to maintain a character's facial identity with several i2v models or through i2i techniques, not to mention since this is a nobody character, there's no recognition affinity, so then why add to the wash of AI characters already out there?

well... since this is a Klein model, you can use transpose characteristics very easily with the edit feature.

Want to add a consistent BBC on an existing image for fun and giggles?

Edit Prompt: replace his penis <lora:this_lora>

Bam! (Mr. Biggz?)..

Want to transform a lanky twink to a muscle bound Gigachad with a consistent body type?

Edit Prompt: transform his body <lora:this_lora>

Shabazz .. er, Shazam!