Originally posted to HuggingFace by TopdeckingLands
Art of MtG v1
Experimenal model, based on ~5000 arts for cards from Magic: the Gathering game on top of SD1.5 model. Large variety of cards from 2014 to 2022 was used. Logos, guild/clan icons and mana symbols were not part of training set as Fan Content Policy explicitly prohibits their use. Each art was captioned with card name, artist, colors and name of color combination, type line, set name and BLIP description.
You can try it out on Huggingface Space (probably slow) and Google Colab
Usage
mtg art is the main prompt to use, and adding card frame negative prompt is recommended.
Model has understanding of some planes, terms and characters to different degree. Previews below use negative prompt lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, Euler sampler, 20 steps and CFG 7.
Planes
Planes that did not have sets named after them are unknown to model.
Characters
Even main cast of planeswalkers that appear often does not work consistently, even if it's often possible to recognize character - the styles are wildly different between specific original arts.
Species
Model has good understanding of phyrexian and Emrakul(tentacle) type of eldrazi among others
Artists
A total of 200 artists had at least 5 images included, 50 of them had more than 30 each. It's safe to expect every major modern MtG artist style to be represented and reproducible to some degree. Styles are not supposed to be representative of artist's real style and are biased by types and colors of cards illustrated by them.
Color combination names used
Different color combinations have different aesthetic attached. X room for example:
Improvements considered for V2
Fine-tuning data was provided by Wizards of the Coast under their Fan Content Policy. Note you can't use this model for commercial purposes, as it's strongly against given policy.
Art of Magic the Gathering is a highly specialized Image generation AI Model of type Safetensors / Checkpoint AI Model created by AI community user civitai. Derived from the powerful Stable Diffusion (SD 1.5) model, Art of Magic the Gathering 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 Art of Magic the Gathering is capable of generating images that are highly relevant to the specific use-cases it was designed for, such as style, fantasy.
With a rating of 0 and over 0 ratings, Art of Magic the Gathering is a popular choice among users for generating high-quality images from text prompts.
Yes! You can download the latest version of Art of Magic the Gathering from here.
To use Art of Magic the Gathering, 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 Art of Magic the Gathering, check out our crash course in AI image generation.
Go ahead and upload yours!
Your query returned no results – please try removing some filters or trying a different term.