README.md

---
pipeline_tag: image-to-image
tags:
- image-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- output:
    url: images/example_001.png
  text: 'Next Scene:  The woman smokes a cigarette, dramatically.'
- output:
    url: images/example_002.png
  text: 'Next Scene:  The woman walks towards a man in the car'
- output:
    url: images/example_003.png
  text: 'Next Scene:  The cow is eating grass in the fields'
- output:
    url: images/example_004.png
  text: 'Next Scene:  The man is shaking hands with another man.'
- output:
    url: images/example_005.png
  text: 'Next Scene:  An old hand puts a glass of wine on the table next to the book.'
license: apache-2.0
---

# Next Scene Lora

<Gallery />

## Model description

This is an image-editing LoRA. Provide an input image and an instruction describing the edit. **Trigger word:** use `Next Scene: ` in your instructions for best results. The examples below show input → output for different instructions.

## Trigger word

Include **`Next Scene: `** in your edit instruction, for example:

```
Next Scene:  <your edit instruction here>
```

## Gallery

### Example 1

![Example 1](images/example_001.png)

*Prompt:* Next Scene:  The woman smokes a cigarette, dramatically.

*Input:* ![Input 1](images/input_001.png)

### Example 2

![Example 2](images/example_002.png)

*Prompt:* Next Scene:  The woman walks towards a man in the car

*Input:* ![Input 2](images/input_002.png)

### Example 3

![Example 3](images/example_003.png)

*Prompt:* Next Scene:  The cow is eating grass in the fields

*Input:* ![Input 3](images/input_003.png)

### Example 4

![Example 4](images/example_004.png)

*Prompt:* Next Scene:  The man is shaking hands with another man.

*Input:* ![Input 4](images/input_004.png)

### Example 5

![Example 5](images/example_005.png)

*Prompt:* Next Scene:  An old hand puts a glass of wine on the table next to the book.

*Input:* ![Input 5](images/input_005.png)

## LoRA scale

The gallery was generated with **`lora_scale=2.0`**. 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-edit-lora:17651bd22e8c151cdb13a97b0f8554dce1e7238cd0a18cf90bc237ac5f0bc067",
    input={
        "images": ["https://example.com/input.png"],
        "prompt": "Next Scene:  your edit instruction",
        "lora_weights": "https://huggingface.co/davidberenstein1957/p-image-edit-next-scene-lora/resolve/main/weights.safetensors",
        "lora_scale": 2.0,
        "hf_api_token": "your-hf-token",
    }
)
```