AI OCR
Extract Text from Image is for reading text from images, scans, and PDFs. Use it for invoices, forms, and other documents you need as text for the rest of the scenario.
Plug and Play (PnP) billing
This is a PnP (Plug and Play) node: Latenode meters usage and charges PnP tokens on top of execution credits (1 PnP token = $1). This node has custom pricing: check the node settings for the current rate.
Fields
| Field | Description |
|---|---|
| User Prompt | Message to the model. Default: "Hello! OCR this image." At least one of User Prompt or Dialogue History JSON is required |
| Attachments (required) | Key-value pairs. In Key, the file name with extension (from the data pop-up, e.g. 1.body.files.[0].filename, or typed manually, e.g. example.jpg). In Value, a file URL (e.g. https://example.jpg) or file content from the data pop-up. Accepted formats: .jpeg, .jpg, .png, .webp, .pdf (only one PDF document per run) |
| Page Number for OCR | The page of the document to process when Attachments contains a PDF |
| Scale Document | Page scale, 0 to 7 (e.g. 1.5). Decrease if the response shows "finish_reason": "length"; increase if text isn't recognized due to poor quality |
Limitations
The operation may work incorrectly if more than 1 file content value and more than 3 file URLs are provided. Only one page of a PDF document is processed per run: use Page Number for OCR to pick the page. If the service can't recognize text on the page, use Scale Document.
Mistral OCR
Extract text and structured data from PDFs with the Mistral OCR model, PnP billing.
Perplexity
Use Perplexity AI search models and web search in Latenode scenarios with built-in PnP pricing.
Need Help? Ask the community
If something on this page is missing or unclear, post on the Latenode community forum. Our team and other users usually reply quickly.