-This is not a model specificly made for porn, it is
Ever want to prompt for something a bit more specific? Now you can. This model was trained with only high res images, of women from various ethnicities. I then gave each their own tokens so you can recall them.
This model is unique in that I actually tried really hard to hand select the images, and there are a lot. The dataset is around 8400. I chose many models, and used about 99% photoshoot images.This model is NOT for making dolls. This is for making girls that look more like real people. This is not going to give you that fake cgi look. Like most realism models, it works best for closeup stuff, and will start to break as you zoom out.
Now, what Can this do? Mostly anything. It was trained on a mix of nude and clothed images, but there are no porn images used. This model mixes really well with anime models to create semi realism looks.
Atm the options are: Caucasian, Korean, Japanese, Chinese, Ebony, Desi
caucasian, european
koran, korea
chinese, china,
ebony, African
Desi, indian
They also have lunachan as a tag, if you just want to make the realism better but don't want to choose. Using dark skin tag will help if you mix this with others, it will recall this and make the skin darker.
Fix Ebony, I have no clue why its not working, like at all...
Maybe try to add one more ethnicity, but would need suggestions on what the most apt choice would be, and it would need to be findable.
This is a merge of my lora and chillmix- I tried several options, and this seemed to be the best product overall. Will release the lora when I get a chance.
Luna's Angels- A realism model with Ethnicity support is a highly specialized Image generation AI Model of type Safetensors / Checkpoint AI Model created by AI community user lunachan. Derived from the powerful Stable Diffusion (SD 1.5) model, Luna's Angels- A realism model with Ethnicity support has undergone an extensive fine-tuning process, leveraging the power of a dataset consisting of images generated by other AI models or user-contributed data. This fine-tuning process ensures that Luna's Angels- A realism model with Ethnicity support is capable of generating images that are highly relevant to the specific use-cases it was designed for, such as girl, person, photorealistic.
With a rating of 4.33 and over 9 ratings, Luna's Angels- A realism model with Ethnicity support is a popular choice among users for generating high-quality images from text prompts.
Yes! You can download the latest version of Luna's Angels- A realism model with Ethnicity support from here.
To use Luna's Angels- A realism model with Ethnicity support, download the model checkpoint file and set up an UI for running Stable Diffusion models (for example, AUTOMATIC1111). Then, provide the model with a detailed text prompt to generate an image. Experiment with different prompts and settings to achieve the desired results. If this sounds a bit complicated, check out our initial guide to Stable Diffusion – it might be of help. And if you really want to dive deep into AI image generation and understand how set up AUTOMATIC1111 to use Safetensors / Checkpoint AI Models like Luna's Angels- A realism model with Ethnicity support, check out our crash course in AI image generation.
You can now Prompt Desi!
Ebony has a good dataset now!
Chinese dataset has been improved, they are now all equal in dataset size give or take!
Also something broke with ebony, and you likely need to use dark skin to make it work. I have looked everywhere but can't figure out why this has happened. It makes white skin... but like, there are no ebony tags in anything but it. African tag also doesn't work, so straight up don't get it. And I tried 3 trainings and none work. Because ebony is broken there is a possibility it is bleeding Desi into its generations, I'm unsure if this is true, but just possible.
Training steps seems lower, but I used a higher batch. So this was overall trained longer.
Go ahead and upload yours!
Your query returned no results – please try removing some filters or trying a different term.