Tips for training Firefly custom models

Training essentials

• Choose at least 10 - 30 high-quality images that showcase the styles and concepts you want the model to generate.
• Include various viewpoints and backgrounds while maintaining a consistent aesthetic.
  
• Crop your sample images to focus on the most important visual elements.
  
• Ensure the images have a resolution higher than 1024 x 1024 pixels with a maximum 16:9 aspect ratio for landscape or 9:16 for portrait. 
  
• Use JPG or PNG files.

Model-specific best practices

Subject models (tech preview)

Custom models trained on a subject, such as objects, characters, or products, will identify key features of the subject and attempt to replicate it in different positions and environments.

• Object consistency: Provide images of the same make and model as your subject. Avoid mixing multiple colors and ensure a common theme or pattern among images for effective learning.
• Object focus: Use images of the subject in clear focus without unnecessary distractions. The subject should occupy at least 25% of the image's area.
  
• Environmental context: Provide images of the subject in different views and contexts, showing it in a variety of lighting conditions. While images with white or transparent backgrounds can be used, it's best to have a mix with more complex surroundings as well.

Style models

Custom models trained on a style will identify the look and feel of the assets to generate similar images when prompted.

• Provide similar aesthetics: Include images that show various scenes and objects while maintaining the same look and feel.
• Use various images: Use as many images as you can to prevent the model from focusing too much on unwanted objects or subjects.
  

Helpful resources

• Train Firefly custom models
• Create on-brand content with Adobe Firefly
• Writing effective prompts
• Get help with Firefly for enterprise