Caption Image
Caption Image
Section titled “Caption Image”Describe an image in a short sentence, running locally in your browser (ViT-GPT2 on transformers.js). Private, offline, no cost.
What it does
Section titled “What it does”Runs the image through an on-device image-to-text model and returns a natural-language caption.
When to use it
Section titled “When to use it”Generate alt text, summarize an image for a downstream text step, or make a searchable description of visual content — without a vision API.
Inputs and settings
Section titled “Inputs and settings”| Setting | Notes |
|---|---|
| Model | Any installed image-captioning (image-to-text) model. |
| Image | Image URL, data URL, or $binary from a previous node. |
Outputs
Section titled “Outputs”Returns { text } — the generated caption string.
Troubleshooting
Section titled “Troubleshooting”- Empty or generic caption — captioning models are small; keep expectations modest, or feed a clearer image.
- Empty picker — install an image-to-text model first.