Jump to content

AI illustrations of DOAXVV chara.

Guest

By Guest,
in Dead or Alive Xtreme: Venus Vacation

Recommended Posts

callum_callum

callum_callum

Posted
Posted

I would like to try generating pictures like these but I have no idea how to start. Most online AI picture generators are paid.  Can somebody create a very short guide on how to begin, what online generators to use, or what software is the best?   Do you generate this just by typing prompts or does it require more specialised knowledge? Any help would be appreciated. Importantly, I don't want to use any paid sotfware/websites.

himawari56

himawari56

Posted
Posted (edited)
13 hours ago, callum_callum said:

I would like to try generating pictures like these but I have no idea how to start. Most online AI picture generators are paid.  Can somebody create a very short guide on how to begin, what online generators to use, or what software is the best?   Do you generate this just by typing prompts or does it require more specialised knowledge? Any help would be appreciated. Importantly, I don't want to use any paid sotfware/websites.

You can use the website TensorArt. Free users have their generation points reset by 50 points daily. Generating two 1024x1536 images costs 3 points of computing power each time.
For character generation, you can choose the Illustrous checkpoint and Lora.
For DOAXVV characters, choose the checkpoint suitable for 3D characters.
This is the checkpoint that currently generates the best results.

This website currently prohibits NSFW content, but you can cheat by switching the number of lines. For example, if you have the line 1girl, 1boy, standing, sex from behind, penis, kiss from behind,
puffy nipples, bedroom,
the system will prompt "xxx is not permitted. See our updates on policy."
You can modify it to the following version, placing each sensitive word on a separate line:
1girl, 1boy, standing, kiss from behind,
sex from behind
penis
puffy nipples
bedroom,
The key point of this modification is that there must be no spaces or symbols before or after any sensitive words, otherwise they will be detected by the system. There are some exceptions, such as "sex from behind" and certain sensitive terms.

You may need to test this yourself.

Also, the site's search function is quite poor, making it difficult to find the Lora you're looking for. In this case, try searching for the site name and the Lora you're looking for using Google.

If you have further questions, you can join their discord and someone will be able to help you.

Edited by himawari56
1
huehuehue9001

huehuehue9001

Posted
Posted (edited)
On 10/3/2025 at 10:34 PM, callum_callum said:

I would like to try generating pictures like these but I have no idea how to start. Most online AI picture generators are paid.  Can somebody create a very short guide on how to begin, what online generators to use, or what software is the best?   Do you generate this just by typing prompts or does it require more specialised knowledge? Any help would be appreciated. Importantly, I don't want to use any paid sotfware/websites.

 

Google 4chan /edg/ - Ecchi Diffusion General

You can generate locally.

Just need a gpu. 8gb vram minimum. 16GB recommended.

Nvidia recommended because CUDA.

AMD will need some setup.

RTX 3060 12GB for cheapest.

4060ti / 5060ti 16GB for cheapest 16GB.

You will need storage space on your SSD. It grew and grew now mine is 350GB. I gave up managing and bought another nvme ssd. Thankfully SSD is cheap now. You don't have to buy fancy fast ones.

 

You can text to image (t2i) and image to image (i2i).

t2i is recommended to generate at 1024x1024 or 1MP because they are trained at the resolution. You can go higher but you will get higher errors, deformed body.

That is where i2i comes in you use the t2i image. feed it in i2i to basically redraw on top at a higher resolution.

 

https://github.com/easydiffusion/easydiffusion#installation

I never tried this but 4chin said it's easy.

 

I installed Auto1111.

When I look back at it, It's pretty bad, but it works.

Then I installed Comfy UI and gained understanding of how the thing works.

It can be scary but actually not. Comfy already provide basic templates that is easy to understand.

 

 

Your end goal should be Comfy UI.

You will need this if you want to do video or any other AI.

The diffusion thing is only stable diffusion, for still image.

Comfy UI is also better in memory management.

 

What I would recommend?

Probably Forge or ReForge for a month. Then go get Comfy UI.

 

comfy UI guide. here.

https://www.youtube.com/watch?v=g74Cq9Ip2ik

 

After installing, you should go to civitai and download WAI NSFW model.

you can try other models as well.

the stock stable diffusion model is not trained on anime.

Edited by huehuehue9001
1

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...