If interested, get in touch for professional services and highly personalized models for commercial purposes.
Please note, we are dealing with a high volume of requests.
Join our DISCORD
from Team Crystal Clear
How to use
Tested in Automatic1111:
Sampling Method: DPM2, Heun, Euler
Schedule Type: Uniform, SGM Uniform, Simple, Normal
DPM++2M Simple
Tested CFGs: 2 to 8
COMFYUI
Load the clips manually, no VAE, try same samplers as above
I will post more feedback on this.
SD3 - Medium
Trained on Stable Diffusion 3 Medium, FP16, we aimed to improve some of the weaker points in SD3. This version was not built to bring drastic changes to quality. With this 1st iteration we aimed to add some artistic concepts: lineart, illustration, flat style, vector images, better body proportions and consistency in full body and medium shots. High number of fashion related items such as cardigans, sweaters, shirts, etc.
This model is free for personal use. For commercial use read below.
If you are to use this model in your merges please give us a much deserved holla.
We do not want to start flagging our content in order to find out if it's being used without our permission, so please, be a team player and let us know.
About Team Crystal Clear
We might not necessarily be known for the amount of content made public on Civitai, but we are known in the industry since we mainly focus on training Stable Diffusion content for commercial purposes.
RunDiffusion, LogoDiffusion, ThinkDiffusion, GenerativeLabs, LunaLustAI, Fanvue, fashion and apparel companies, instagram influencers and models, automakers, and other private and public companies.
Such content is considered confidential and it is at our client's discretion if they want to release it or not.
Please visit our DISCORD for more information and quotations.
We have a team of highly skilled people with access to some of the latest hardware.
If we helped save on your business costs by using this model to generate content for commercial purposes, you may express your gratitude by clicking this DONATE button.
Description
Tokens representative of trained content
Visual Styles:
lineart, illustrative, cartoon, flat style, vector illustration
Various articles of clothing:
shirt, sweater, pants, jeans, hoodie, cardigan, jacket, coat, denim
Auesthetics:
venomized, sharp teeth, rage face
Baked VAE
FAQ
Comments (6)
Don't know if i'm really into the restrictions of SD3, but this is pretty cool.
Without clips right?
I just keep getting this
TypeError: 'NoneType' object is not iterable
SwarmUI is not seeing it as a SD3 model. SwarmUI with load the options for the clip encoders automatically when a SD3 model is selected but with this model it doesn't. I'm assuming the encoders are not trained into the model due to the small size???
Why are the images I'm generating blurry? Can you give me an example of the workflow and settings for the sampler and scheduler?
I couldn't replicate any of the gallery images with this, CFG 7 burns everything, did anyone have any luck? is it just me or does this model not perform? Not seeing any other images posted makes me think it's not just me...
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.