---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- output:
url: images/example_001.png
text: DADADOLL style A figure with geometric shapes and bold colors in a surreal
interior.
- output:
url: images/example_002.png
text: DADADOLL style A portrait with fragmented planes and cubist influences.
- output:
url: images/example_003.png
text: DADADOLL style A still life of fruit and objects rearranged in an abstract
composition.
- output:
url: images/example_004.png
text: DADADOLL style A cityscape with distorted perspective and vivid blocks of
color.
- output:
url: images/example_005.png
text: DADADOLL style A face composed of overlapping geometric forms and sharp angles.
license: apache-2.0
---
# Photos Modernism Art Lora
## Model description
This is a text-to-image LoRA. **Trigger word:** use `DADADOLL style` in your prompts for best results. The gallery above shows sample generations.
## Trigger word
Prepend **`DADADOLL style`** to your prompt, for example:
```
DADADOLL style
```
## Gallery
### Example 1

*Prompt:* DADADOLL style A figure with geometric shapes and bold colors in a surreal interior.
### Example 2

*Prompt:* DADADOLL style A portrait with fragmented planes and cubist influences.
### Example 3

*Prompt:* DADADOLL style A still life of fruit and objects rearranged in an abstract composition.
### Example 4

*Prompt:* DADADOLL style A cityscape with distorted perspective and vivid blocks of color.
### Example 5

*Prompt:* DADADOLL style A face composed of overlapping geometric forms and sharp angles.
## Example prompts
Prompts used for the gallery above:
- DADADOLL style A figure with geometric shapes and bold colors in a surreal interior.
- DADADOLL style A portrait with fragmented planes and cubist influences.
- DADADOLL style A still life of fruit and objects rearranged in an abstract composition.
- DADADOLL style A cityscape with distorted perspective and vivid blocks of color.
- DADADOLL style A face composed of overlapping geometric forms and sharp angles.
## LoRA scale
The gallery was generated with **`lora_scale=0.5`**. You can tune this when running inference (e.g. lower for a subtler effect, higher for a stronger style).
## Use on Replicate
```python
import replicate
output = replicate.run(
"prunaai/p-image-lora:393190dc623ef136356d6a7221c561f48771f39c5cff90ef903ca49d1bd47574",
input={
"prompt": "DADADOLL style your prompt here",
"lora_weights": "https://huggingface.co/davidberenstein1957/p-image-photos-modernism-art-lora/resolve/main/weights.safetensors",
"lora_scale": 0.5,
"hf_api_token": "your-hf-token",
}
)
```