- The media must be fully generated by an AI tool via prompts without any post processing nor resizing.
- The media can be anything that can be show on the screen, it may include just a picture but is not limited to that, audio and video and combinations are allowed as long as they are generated by AI without intervention by a human
- AI upscaling is allowed as part of the generating process as long as it happens automatically and does not require human intervention
- No starting media may be used - no white noise, no AI generated pictures, nothing.
- No in- or out-painting on the image
- You must submit clear evidence of the process of creating the media in the form of at least some versions (appropriate amount depending on the process and tool) of the unfinished media that demonstrate the creation process.
- create a folder that is the tittle of the art piece
- place the art piece into the folder, in the format it came out of the AI tool, do not remove any metadata from the file. name it same as the folder.
- create another folder in this folder called process
- place the prompt for the art into process folder into a text file called prompt.txt
- place screenshots you have taken during the creation into the process folder
- place the information about the system used (for ex. ComfyUI), model checkpoint, LoRA, LyCORIS and any plugins used into a textfile called system_info.txt
- finally compress the first folder into a zip file with the same tittle as before, so it becomes your_artpiece_name.zip so you may submit it as a single file
- The media must be previously unpublished before the competition.