Type: huggingface.multimodal.ImageToText
Namespace: huggingface.multimodal
Description
Generates text descriptions from images. image, text, captioning, vision-language
Use cases:
- Automatic image captioning
- Assisting visually impaired users
- Enhancing image search capabilities
- Generating alt text for web images
Properties
| Property | Type | Description | Default |
|---|---|---|---|
| model | hf.image_to_text |
The model ID to use for image-to-text generation | {'type': 'hf.image_to_text', 'repo_id': '', 'path': None, 'variant': None, 'allow_patterns': None, 'ignore_patterns': None} |
| image | image |
The image to generate text from | {'type': 'image', 'uri': '', 'asset_id': None, 'data': None} |
| max_new_tokens | int |
The maximum number of tokens to generate | 50 |
Outputs
| Output | Type | Description |
|---|---|---|
| output | str |
Metadata
Related Nodes
Browse other nodes in the huggingface.multimodal namespace.