Skip to main content

Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines

You must be 18+ to view this content

Starmaker Story may contain content you must be 18+ to view.

The creator of this page has specified that it contains mature themes and content. Please enter your birthdate to verify you are 18 or older:

or Return to itch.io

i create a Anna LORA

A topic by Jhef_012 created Apr 20, 2026 Views: 9,493 Replies: 29
Viewing posts 1 to 11
(+2)(-1)

I create a own Anna LORA and I want to show you the Image generation, with and without Pixel Art.

Much free time for this xD

I hope you liked it

Pixel Art: image.png

No Pixel Art: image.png (Reduced quality because Itch won’t let me upload it to 4K)

See you!

And sorry for create a LORA, but Anna is just too beautiful. SORRY!!! (xD)

which site you make this?

My PC (Laptop actually) - ComfyUI

(+2)

Nice... now give me romance, give me fantasy... give me post editing and animations... 


My GPU it’s almost dying xD

Let him rest for a while, he need it for the future things.

It looks really great. I’m looking forward to seeing more of your work.

Thanks, I post animations on X every 2-3 days and Jhef_012 posts all his images all at once when ever the wind is blowing east...

Yes. I post random bullshit <3

(+1)

Random gems 😉

(+2)

I guess what I am really trying to say as kindly as possible, creating LoRas is fine but you are only going to end up with cheap imitations of Starmaker CGs and you are not gaining anything from it. 

Try and take time on working on an single image and focusing on finding your own style... I by no means think I am where I want to be yet, but I try and take inspiration from multiple sources when doing my images of Anna. For example with this one keeping Anna's base model and posture from Starmaker but infusing it with other art styles I love as well like Oshi no Ko


 

True. But I’m new on this and I know almost nothing. But thanks for the recommendations, maybe I’ll try, maybe not but it’s great to see your AI Art and animations now. Thnaks man

which site you make this?  

(1 edit) (+2)

It is not a site, well there are some sites used in the process but the majority of work goes on in Stability Matrix/Comfy UI and Photoshop. I have listed below a workflow from an older thread. I have made some tweaks to my own setup since then, but should be enough info to get you going:

My images are currently generated using Stability Matrix with WAI-illustrious-SDXL, SDXL VAE, HiResfix and Yolo for face & hands

I then do a lot of manual  editing in Photoshop for clean up and backgrounds

I am currently using Comfy UI with Wan2.2 for animations (The "Wan 2.2 14B Image to Video" is a simple one to use but you need to relink a couple of nodes for it to work)

Then I use Ezgif to convert MOVs to GIFs 

Then 8Upload for SFW images and Imgbox for NSFW ones

Quick question about all of the images generated by AI to anyone who is nice enough to reply: Why are they almost always clean-shaven? Why is that the standard for the image generation? In hentai, women are hairy too sometimes but it somehow never translates to AI images. In this thread as well, there is no hair. Is it still possible to produce hairy images or is there a technical limitation that I am not aware of? (I honestly don't know anything about AI image generation other than it is based on a training data set that skews the results in certain ways.) 

(+4)

No very possible, with a good person yolo detailer for example, a lot of it is down to personal preference and who the image is based on. Example in game Anna never has pubic hair, and there does not appear to be out cry from the fanbase for her to do so. In porn a bush is normally a niche category, though I am sure there are plenty of prompters and tracers out there fitting this market... 


Just know that I love you for answering AND providing an example image. What a gorgeous picture. My week has been saved by your generosity, may your pillows always be cold! 

(+1)

awsome 🫦🫦..

(+5)

Lol on my phone the image looked like a shitty Itch thumbnail, so I played around with it some more and now I cursing y'all with my imagination 😅


You always do a fascinating work.

Much rather: gracing our eyes with unparalleled, heavenly beauty.

(+2)

Creo que ya le voy agarrando la onda, pero falta mejorar.

image.png

This is great! Can you share the Lora?

Nah, I don’t think so, not for now at least

Can you share the Lora?

Nah, I don’t think so, not for now at least

(+1)

Thats really cool - I just started my own image generator.  How many image/text training sets did you create - what was your sweet spot? 

(+1)

Actually, I used 341 images with captions, but the training was only 1600 steps at 0.0001 LR. Maybe it’s too few steps for such a large dataset, but … it’s working, you know?

Work for poses and NSFW, so for me it’s fine :)

(Pero es una salvajada en términos técnicos xDD)

image.png image.png image.png

(+2)

That's so cool - dang, I got a lot of work to do....I'm using Koyha for the LoRA  and Stable Diffusion for image generation (I'm trying to create realistic image generator).  I have a new found respect for these generators/ai training after trying this - we both know how much work goes into this...How did you caption and verify all 300+ images quickly?  Appreciate the tips and great work.

Actually, I used the built-in WD14 Tagger utility inside kohya_ss. It’s much easier than writing a txt manually for every image, especially for 300+ images, because it generates all the captions automatically. I just added the names ‘Anna’ and ‘Starmaker’ to everything. It’s very basic, but it works, as you can see. 😅