Generating images of women with small breasts using AI image generation models like Stable Diffusion can be challenging. The AI model has likely been trained on datasets with a bias towards larger breast sizes. However, with careful prompt engineering, it is possible to guide the model to generate more petite figures.
In this article, we will explore prompt strategies for reducing breast size in Stable Diffusion images. We will cover both positive and negative prompts, weighting, artistic styles, framing, and other techniques to consistently yield small-breasted results.
Positive Prompts
Here are some positive prompts to try that directly describe small breasts:
- (small breasts)
- (tiny breasts)
- (flat chest)
- (petite breasts)
- (modestly endowed)
Wrap these prompts in parentheses to boost their influence in the final image.
You can also combine multiple prompts, like:
(small breasts), (flat chest), petite
Other helpful descriptors:
- slender
- skinny
- thin
- narrow shoulders
- delicate collarbone
- petite frame
- slim
- lean
Use adjectives like “dainty”, “petite”, “slim”, and “narrow” to describe features besides just the breasts. This guides the model to generate an overall smaller figure.
Negative Prompts
Negative prompts are just as important as positive prompts. Here are some useful negative prompts for reducing breast size:
- (large breasts)
- (huge breasts)
- (big boobs)
- (massive melons)
- (giant jugs)
- (enormous breasts)
- voluptuous
- busty
- top-heavy
- front-heavy
The more exaggerated descriptors in the negative prompt, the more it will suppress those attributes.
Weighting
You can control the influence of prompts by assigning a weight value from 0 to 1. For example:
(small breasts):1.2, busty:-0.8
Higher positive weights and more negative negative weights will steer the output more strongly.
Artistic Style
The choice of art style can also affect breast size:
- anime
- Disney
- Pixar
- cartoon
- illustration
- vector art
These styles naturally lend themselves better to petite figures.
Conversely, avoid art descriptors like:
- pinup
- glamour photography
- boudoir
Framing
Cropping and framing is another useful technique. Close-up portraits and headshots eliminate the issue of breast size altogether.
- portrait
- headshot
- close-up
Full body framing tends to make breast more prominent. Avoid prompts like:
- full body
- head to toe
Advanced Prompts
For best results, combine multiple techniques into one prompt:
((small breasts)):1.2, ((flat chest)), skinny, narrow shoulders, delicate collarbone, ((busty)):-0.8, headshot, Disney style illustration
This advanced prompt includes:
- Multiple positive descriptors of small breasts
- Negative suppression of large breasts
- Weighting to strengthen influence
- Related body descriptors
- Negative framing prompt
- Artistic style prompt
With practice, you’ll develop an intuition for prompts that reliably yield petite results.
Example Images
Here are some examples of Stable Diffusion images generated using prompts for small breasts:
Example 1
Prompt: (small breasts), flat chest, delicate collarbone, portrait
Example 2
Prompt: ((tiny breasts)):1.2, busty:-0.8, Disney style
Conclusion
Generating petite breast sizes requires carefully engineered prompts and negative guidance. But Stable Diffusion offers powerful control through positive and negatives, weights, artistic modifiers, framing, and more. With experimentation, you can learn prompts that work for your desired aesthetic.