Type: huggingface.image_to_image.ImageToImage

Namespace: huggingface.image_to_image

Description

Transforms existing images based on text prompts using AutoPipeline for Image-to-Image. This node automatically detects the appropriate pipeline class based on the model used. image, generation, image-to-image, autopipeline

Use cases:
- Transform existing images with any compatible model (Stable Diffusion, SDXL, Kandinsky, etc.)
- Apply specific styles or concepts to photographs or artwork
- Modify existing images based on text descriptions
- Create variations of existing visual content with automatic pipeline selection

Properties

Property Type Description Default
model hf.image_to_image The HuggingFace model to use for image-to-image generation. {'type': 'hf.image_to_image', 'repo_id': '', 'path': None, 'variant': None, 'allow_patterns': None, 'ignore_patterns': None}
prompt str Text prompt describing the desired image transformation. A beautiful landscape with mountains and a lake at sunset
negative_prompt str Text prompt describing what should not appear in the generated image. ``
image image The input image to transform {'type': 'image', 'uri': '', 'asset_id': None, 'data': None}
strength float Strength of the transformation. Higher values allow for more deviation from the original image. 0.8
num_inference_steps int Number of denoising steps. 25
guidance_scale float Guidance scale for generation. Higher values follow the prompt more closely. 7.5
seed int Seed for the random number generator. Use -1 for a random seed. -1

Outputs

Output Type Description
output image  

Metadata

Browse other nodes in the huggingface.image_to_image namespace.