The quality of your image output is highly dependent on the quality of your prompt.
Below are some tips to maximize the quality of your prompt:
  • Avoid short, non-descriptive prompts. The more detail the better the output!
  • Separate text with commas for best results (e.g. stunning, cinematic lighting, beautiful, 8K photorealistic)
  • Use parentheses to emphasize important descriptions (e.g. male portrait, ((green hair)), sunset)
  • Use style descriptions to enhance the quality of your render (e.g. 8k, lush, cinematic lighting)
  • Use style descriptions to mimic the style of famous artists and/or photographers (e.g. Monet, watercolor; Picasso, abstract; or cyberpunk, Syd Mead)
  • Add negative prompts to improve typical anatomy problems OR Use brackets with words or multiple words within to let AI ignore unnecessary information for your desired output (e.g. ([bad anatomy, extra legs, extra arms, extra fingers, poorly drawn hands, poorly drawn feet, disfigured, out of frame, tiling, bad art, deformed, mutated]).
  • Use points to blend different sections of the prompt. This helps you explore the AI's latent semantic space and generate interesting (and often surprising!) variations (e.g. blue sphere:0.25 red cube:0.75 hybrid).
  • Append a word or phrase with “-“ or “+“ to increase or decrease the amount of something. (e.g. a tall thin man picking apricots+, a tall thin man picking ((apricots))+, a tall thin man ((picking apricots))-, etc).

Now that you know the basics, let's generate some art!