new test model v 1.1 - try it, it is mainly for testing purposes. Let me know in the comments what you think, I can tune it more in case of problems
MAGIFACTORY model is trained to generate t-shirt logos based on text input. The model takes in a text description of the desired t-shirt logo and generates a corresponding image. It is trained on a dataset of t-shirt logos with resolution 512x512px. Ideally use hires-fix rescale to 896x896 ( for not having repeated artifacts ) in automatic1111 WebUI with "Latents antialiased" and finally upscale 4x with SwinIR_4x. Euler a scheduler preferred. You can also try DPM++ 2S a Karras scheduler. I am running Instagram with top designs, so you can download and use them.
prompts:
cyberpunk astronaut flying in space (on a black background:1.4) as a t-shirt logo in the style of <MAGIFACTORY> arta symbol of a car (on a black background:1.2) as t-shirt logo in the style of <MAGIFACTORY> artnegative prompt:
badly drawn, noisy, noise, blurred, ugly, cropped, out of frame, (double:1.4), (repeated:1.4), (repeating:1.4), signature, text, font, watermarkDescription
Cleaned training images, much less generated text
FAQ
Comments (14)
Could you please add a pickletensor?
Or you can convert it using https://github.com/diStyApps/Safe-and-Stable-Ckpt2Safetensors-Conversion-Tool-GUI
@odawgthat aight thanks
@LAIM_ART :)
LOVE the images the model creates. It's a BIG shame the license restrictions. It REALLY defeats the WHOLE purpose of posting this models. Please re-consider your license, or better, put it behind a paywall. Otherwise, the whole free and open-source does not make much sense. L&L
I am not strictly close minded, we can talk about this. But tell me ... How does restricting the sale of images go against the principles of open source? I believe that the essence of the open source movement is to have access to free and openly available resources for personal purposes. However, I have observed some indie developers who are using models and code created by others to make significant profits without giving back anything to the community, particularly with avatar creations. My aim was to prevent this from happening, while still allowing everyday individuals to use these resources for their personal projects or even for selling t-shirts.
@magifactory I understand your point. I am trying myself to make a side income creating designs for sell. However, I plan to share part of my success (if anything happens) to the creators that enable things to happen. I guess that there's a big difference between individuals like me and businesses.
At this point, I would not use your model to sell T-shirts as the license does not allow it. I really like the results using it but I do not want to get into any trouble that takes more time and effort to deal with it that any small financial gain.
Thank you for your time to get back. L&L
@aferlo78 Ok, I just made more loose licensing. Is it fine for you now?
@magifactory @magifactory
Hi,
Thank you very much for the changes to the license and allow to sell the images generated with your model.
I am sure that more people would also appreciate it.
I am looking forward to create images with your model.
L&L
@aferlo78 do you think that we can talk for 15-20 minutes? I want to know how people use the model
@magifactory Sorry for not getting in touch before. I have been working flat out. I would not know where to fit time currently. However, my intention is to create images that have a value to be able to sell on Etsy as digital download. At the moment, I have selected another niche and I am developing the idea I had. It's a long shot but worth taking it for me.
Best
Was this trained on the base model?
nn, on this ... https://huggingface.co/22h/vintedois-diffusion-v0-1
Thank you for all you hard work and effort into making this model
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.