Learn PromptingQuality Boosters
- Quality Boosters are terms that enhance style-agnostic attributes, such as detail and resolution, in AI-generated images.
What are Quality Boosters?
Quality boosters are terms added to a prompt to improve certain non-style-specific qualities of the generated image. For example "amazing", "beautiful", and "good quality" are all quality boosters that can be used to improve the quality of the generated image.
An Example of Quality Boosters
Recall from the other page the pyramids generated with DALLE, and the prompt pyramid.
Now take at pyramids generated with this prompt:
Prompt
These are much more scenic and impressive!
Here is a list of a number of quality boosters:
High resolution, 2K, 4K, 8K, clear, good lighting, detailed, extremely detailed, sharp focus, intricate, beautiful, realistic+++, complementary colors, high quality, hyper-detailed, masterpiece, best quality, art station, stunning
Conclusion
Quality boosters, like style modifiers, demonstrate how small updates to an image prompt can lead to drastic improvements in generated outputs.
FAQ
How can quality boosters improve my image prompts?
Quality boosters can improve the non-style-specific qualities of an AI-generated image. As shown in this article's example, adding terms like "beautiful," "majestic," "incredible," and "4k" to our prompt leads to noticeable differences in the quality of the outputted image of a pyramid.
Notes
Similar to the note on the previous page, our working definition of quality boosters differs from Oppenlaender et al.. This being said, it is sometimes difficult to exactly distinguish between quality boosters and style modifiers.
Sander Schulhoff
Sander Schulhoff is the CEO of HackAPrompt and Learn Prompting. He created the first Prompt Engineering guide on the internet, two months before ChatGPT was released, which has taught 3 million people how to prompt ChatGPT. He also partnered with OpenAI to run the first AI Red Teaming competition, HackAPrompt, which was 2x larger than the White House's subsequent AI Red Teaming competition. Today, HackAPrompt partners with the Frontier AI labs to produce research that makes their models more secure. Sander's background is in Natural Language Processing and deep reinforcement learning. He recently led the team behind The Prompt Report, the most comprehensive study of prompt engineering ever done. This 76-page survey, co-authored with OpenAI, Microsoft, Google, Princeton, Stanford, and other leading institutions, analyzed 1,500+ academic papers and covered 200+ prompting techniques.