Donkey-dong twink! Get your donkey-dong twink right here folks!
I'm having too much fun w/ Flux 2 Klein - being able to edit features w/ generation means I can create character + concept based LoRAs into one. I was able to use a previous dataset used for one of my consistent cock concept LoRAs and mesh that in with an AI generated character set making this one.
Who is this guy? We'll he's Kris Strait of course, duh! Heh.. it's of course a Nobody AI character so they only exist here in this debut. I based this one off a twink-bod model using an additional AI generated face dataset at random to see if I can easily Frankenstein a new consistent character with um, a particular attribute. Before I was having to create a full character and then creating the datasets that way, which is more time consuming.
So why did I do that? Well, besides just adding to the plethora of AI characters out there, since this is a Klein LoRA, this can be used to spot-edit and create a specific large uncut weenie on any existing photo with ease. Previously, I was creating either a character based lora and/or a concept based LoRA and then either generating then inpainting or hoping that combining the two during the generation works out.
Want to make that add a monster uncut schmeat to that boring photograph of a bronze statue?
Edit prompt: replace his penis <lora:this lora.safetensors>
.. and boom, now he has a 8 incher uncut donkey-dong.
Want to make that one photo with a guy sporting a dad-bod look like he did back in his early twenties?
Edit Prompt: replace his body and torso <lora:this lora.safetensors>
So even if twink bod guys aren't really your thing, this still can be of use for peen edit/generation purposes.
Of course you can also use this to straight-up generate the full character if that's what you want too.
Enjoy.
Description
FAQ
Comments (3)
So, I'm confused. How do you trigger the LoRA? This phrase "replace his penis <lora:this lora.safetensors>" confused me and I didn't see it in the prompts of the gallery examples. Thanks for your guidance. The world needs more donkey dicks.
Flux Klein has an edit image mode ability (Like Flux Kontext) - it works on existing images. So if you use it that way with an existing image using that functionality then you give it contextual instructions like 'change his hat color from blue to red', 'change the background to a crowded wine bar' etc etc. Check YouTube on how to get some more creative examples and how to set it up with your particular AI image generation platform - I use Krita for photo editing using a plugin that hooks into Comfy as a backend and for me it's just a drop down menu item to move from regular image generation to edit mode. I think there's a few articles on here on how to do it with Comfy's UI as well.
Also - this LoRA isn't can generate full images the regular way and not just used for photo editing. The other gallery images show examples of that.