This model is best at producing landscapes and art pieces rather than game art or stand alone characters.

Trained on 122 1k images where each 'pixel' is 4x4 pixels.

Trigger word:

pixel art

I added another version of the model 1.1 game_art_included. This dataset has more game art style map images. To get a good result add these tags: "top down" "game art" "tile map"

Additional tags for perspective:

top down

isometric

side perspective