MASC [Krea 2]
Photorealistic, diverse men in an editorial style: strong facial structure, realistic skin, and dramatic light. Co-developed by Divergent AI and Gen Hunks.
Base model: Krea 2
Trigger words: none. Write natural language and start with the subject.
Strength: 0.9 recommended (range 0.85–1.0). The version specs list 0.8–1.0.
Resolution: trained at 512. Tested up to 1024px in multiple aspect ratios.
Strengths: masculine faces (jawlines, stubble), skin texture (pores, chest hair, sweat and water), muscle definition under directional light, and open or wet fabrics. It also works for surreal and artistic scenes.
Dataset: 300+ curated editorial, social and artistic semi-clothed portraits. No frontal nudity in the dataset.
Description
Base model: Krea 2
Strength: 0.8–1.0
Resolution: trained at 512, generate at 1024–1536
CFG: 3.5 (Base) / 1.0 (Turbo)
Steps: 35–50 (Base) / 8 (Turbo)
FAQ
Comments (9)
Wow, thanks for sharing this. Male variation is so much needed. This will be very usefull!
Only a pleasure. It's exactly why I started this journey with this LoRA 2 years ago with Flux.
Enjoy it. Let me know if it underrepresents anything. Enjoy!
I would love a gender slider, tried it on women,doesn't seem work sadly.
I was hoping I could get an androgynous look with it.
Hey EventHorizon, as this is a style lora fixated specifically to masculinity it would be very bad at that. There's literally no femininity or androgyny anywhere in the dataset. Only manliness. Hairy chests, beards etc. etc.
For a more generalized model take a look at either of my Vision Vanguard models. They should be good at generating more diverse concepts like androgyny etc.
https://civitai.com/models/2740482/vision-vanguard-25-krea-2
https://civitai.com/models/2741126/vision-vanguard-24-krea-2
I hare the gigantic bs ai description, really. Just be blunt and say what it is, at least in the first few lines.
No, do your LoRA's your way, I'll do mine mine.
Perhaps I’m just too slow on the uptake, but I still can’t understand what it’s for.
It's trained on a dataset of 1000 images of masculine men. Great for getting rid of general base models bias towards feminine content. Makes men generally more hairy and I find generally more appealing to the masculine gaze. Also find it's better at sexualizing men in a softcore manner if you're into that. See it as more of a generalist than a specific focus such as hairy or leather.
It's trained on a bit of everything and I'll expand it as I go.
Not trained on any nudity. Check out my model Manrotica for that.
it makes much better body hair and better sweat, that's for sure.