File size: 1,552 Bytes
30bdfb3 de0ff2d 30bdfb3 7582023 de0ff2d e8a32c6 97c0bd7 3adc008 97c0bd7 a897b62 3adc008 2b64480 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 |
---
title: README
emoji: 🏆
colorFrom: blue
colorTo: red
sdk: static
pinned: false
license: mit
short_description: Master Open Sourced AI, We make Datasets and rarely a model.
---
# Master Open Sourced AI
This is a Group of people who make Ai stuff... Mainly datasets or Stuff like TTI (text to image) models.
We care about Both Open and uncensored AI, since Censoring AI, despite it's benefites for getting rid of bad content can make AIs less straight forwards.
Basically this is only for posting good datasets I've made and so I can show off the models which are the best I've made.
## Current WIPs:
Digital Art dataset: 100-150GB with 30000-50000~ Images.
G.A.R.F.ELID Dataset: 100-200GB with a lot of images
## ABOUT OUR PROJECTS:
If the project has WIP it is not done.
### General Anime,Realistic and Furry Art Dataset or GARFEILD Dataset
Why did we name it Garfeild? Well G for General, A for Anime, R for Realistic and F for Furry, yeah the name sucks but it is better than GARF lol.
Basically the reason why this dataset is being made is for models like SD3 which could create NSFW art BUT they do not know what NSFW art is: Basically just removing data that could tell them how to make NSFW stuff.
Also Well With out a DETAILED knowledge of the human body the best works either made by people or AI cannot be made:
You know since Ai was made to simulate humans and humans usually need a detailed understanding of the human body to create perfect anatomy.
## GARFEILD IS NOT A PRETRAINING DATASET, IT IS A REFINEMENT DATASET. |