ocr_pdf
Extract text from local PDFs and images in 100+ languages, or save a searchable PDF with a text layer.
Instructions
The same tool as ocr_file, kept under its earlier name. Prefer ocr_file.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| path | Yes | Absolute path to a local PDF, PNG, JPG, JPEG, GIF, WebP or BMP file. | |
| output | No | 'text' returns the extracted text; 'pdf' saves a searchable PDF with a text layer. | text |
| output_path | No | Optional absolute path for the result file. An existing file there is replaced. |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |