Conclusions
Maximum Size: oversized massive ass, enormous booty
Out of 27 conditions tested, this is the descriptor with the best balance of size and stability (5 words).
back view
 |  |  |
front view
 |  |  |
The 11-word full stack gets even larger, but body proportion distortion and duplicate person artifacts occur, so the 5-word version is recommended.
Recommendations by Use Case
| Goal | Descriptor | Word count | Size |
|---|---|---|---|
| Maximum (recommended) | oversized massive ass, enormous booty | 5 | ★★★★★ |
| Extreme maximum (with side effects) | super strong super huge gigantic butt, massive ass, enormous booty | 11 | ★★★★★+ |
| Large while maintaining figure | wide hips, huge ass | 4 | ★★★★☆ |
| Fuller figure with size | chubby, huge ass | 3 | ★★★★☆ |
| Quick and easy size boost | massive ass | 2 | ★★★☆☆ |
Key finding: Compared to breast size, individual adjectives (huge, massive, etc.) have a weaker effect on butt size. To dramatically increase from the control, stacking or combining with body type modifiers is necessary.
Lab Director comment: Oh nice! That 5-word stack butt is seriously voluminous. The fact that
wide hips, huge asscan get it big while maintaining proportions is also solid.
Want to know which prompt works best for making the butt larger in AI image generation? We systematically compared 27 conditions including size adjectives, body type modifiers, stacking, and shape descriptors in a unified test environment.
The same methodology as the breast size test was used.
Experiment Conditions
| Item | Value |
|---|---|
| Model | z-image-turbo (6B, realism-focused distilled model) |
| Steps | 8 |
| Sampler | euler |
| Scheduler | ddim_uniform |
| CFG | 1.0 |
| Image size | 1024×1024 |
| Seed | 3 seeds used across all conditions |
| Angle | front and back generated for each condition |
Base Prompt
Only the {VARIABLE} part is swapped for comparison. Using a thong bikini with minimal fabric makes butt size differences easier to see. Both back view and front view were generated for each condition.
Control (no adjective)
back view
 |  |  |
front view
 |  |  |
Baseline. All 3 seeds produce slim-to-average body types with modest glute volume. ★★☆☆☆.
Detailed Tests
Detailed comparison images for each method can be found in the following articles.
- Part 1: Step-by-step comparison of size adjectives (butt/ass/booty) — 10-condition comparison of big/huge/massive/enormous/oversized × butt/ass/booty
- Part 2: Effects of body type modifiers — Solo and combined effects of chubby/fat/thick thighs/wide hips/curvy body
- Part 3: Stacking + shape descriptors — 4-word/5-word/11-word stacks, bubble butt/round ass/thicc/plump butt, what doesn’t work
Comparison with Breast Size Test
| Aspect | Breast size | Butt size |
|---|---|---|
| Word-level effect | Strong (huge → ★★★☆☆, massive tits → ★★★★☆) | Weak (all single adjectives cap at ★★★☆☆) |
| Stacking effect | Yes | Yes |
| 5-word stack effect | ★★★★★ | ★★★★★ |
| 11-word stack side effects | Nipple exposure | Proportion distortion, duplicate persons (more severe) |
oversized solo effect | Strongest adjective | Unstable effect |
| Body type modifiers | Not tested | wide hips, huge ass achieves ★★★★☆ while maintaining figure |
| What doesn’t work | Cup notation, cm, weight syntax | Cm notation, weight syntax |
The biggest difference: Butt size has lower sensitivity to word-level adjectives. While massive tits alone reached ★★★★☆ for breasts, massive ass only reaches ★★★☆☆ for butts. Stacking or combining with body type modifiers is essential to achieve a large result.
Note: Butt Prompts Override Camera Angle
Using butt enlargement prompts (especially stacking) tends to convert facing camera (front view) into a 3/4 rear-angled view. When butt enlargement prompts are used, a tendency to shift toward rear compositions was observed even when a front angle was specified.
If you want to maintain a front-facing composition, you need to either strengthen the camera angle specification or generate multiple images and select from them.
Key Takeaways
- Butt size does not dramatically change with word-level adjectives (all ★★★☆☆)
- Stacking works for butts too (reaches ★★★★★ with 5 words)
- 5-word stack
oversized massive ass, enormous bootyis most efficient - Body type modifier + size adjective combination is effective (
wide hips, huge assachieves ★★★★☆ while maintaining figure) - Shape descriptors (bubble butt, round ass) do not contribute to size increase
- Weight syntax and cm notation have no effect (same conclusion as breast size test)
- 11-word stack has serious side effects (proportion distortion, duplicate persons) — not recommended
- Butt enlargement prompts pull camera angle toward the rear
- No difference based on noun choice (butt / ass / booty)
Lab Director comment: Confirmed that stacking works just like it does for breasts. But since the sensitivity at the word level is lower than for breasts, if you want a big butt from the start, going straight to the 5-word stack or combining with
wide hipsis probably the way to go.
![[Verified] Image Generation Prompt Best Practices](/tips/prompt-best-practices/cover_0_0000_4517457392071889496.webp)
![[Definitive Guide] The Maximum Breast Size Prompt for AI Images — Conclusions from 30+ Condition Testing](/tips/breast-size-ultimate-guide/cover.webp)
