What does it cost to convert PNG to plain text?

$0.005 per call, paid in USDC on Base via x402.

How large can the PNG file be?

Up to 10 MB (10,485,760 bytes). Past a limit the call answers too_large (413).

What happens if the conversion fails?

The tool answers with a typed JSON error from the errors table. Payment is settled only after the tool has produced its result; a call that fails inside the tool is never settled.

catalogue / tools / Documents / PNG to plain text

Convert PNG to plain text (OCR)

To convert PNG to plain text (OCR), send one POST to /v1/ocr-image-to-text with the PNG source. Each call costs $0.005 in USDC on Base via x402, with no account and no API key. The source file can be up to 10 MB.

Endpoint: POST /v1/ocr-image-to-text
Price: $0.005 per call
Service: ocr-image-to-text
Docs: /docs/ocr-image-to-text

01Limits

Limit	Value
Largest source file	10 MB (10,485,760 bytes)
Timeout	45 s
Source formats	GIF, JPEG, PNG, TIFF, WebP
Output formats	plain text
Option lang	one of "eng"; default "eng"

02Example

A real PNG to plain text run of ocr-image-to-text.

Request

POST /v1/ocr-image-to-text
content-type: application/json
payment-signature: <base64 x402 payload>

{
  "url": "https://grist.tools/samples/images/ocr-scan.png",
  "lang": "eng"
}

Response

200 OK
payment-response: <base64 settlement receipt>

{
  "source_url": "https://grist.tools/samples/images/ocr-scan.png",
  "lang": "eng",
  "image_format": "png",
  "text": "SAMPLE SCAN FOR OCR\nSECOND LINE 2026\n",
  "confidence": 86
}

Source file: images/ocr-scan.png, 7 KB (7,168 bytes). 1-bit black-on-white PNG with two uppercase lines drawn from a hand-written 5x7 bitmap font: SAMPLE SCAN FOR OCR and SECOND LINE 2026. The request sets lang to "eng". The recorded run returned 37 bytes of plain text, 1% of the source size, with a mean OCR confidence of 86 and language eng.

In the sample itself: its IHDR chunk declares 556 by 100 pixels, 1-bit greyscale and no interlacing. The chunks run IHDR, IDAT and IEND. The returned text runs to 3 lines, 1 blank one among them, and its longest line, "SAMPLE SCAN FOR OCR", holds 19 characters. Its first line reads "SAMPLE SCAN FOR OCR" and its last "SECOND LINE 2026". Split on whitespace, it yields 7 tokens over 2 non-blank lines.

03Format notes

PNG as a source

Every PNG opens with the eight-byte signature 89 50 4E 47 0D 0A 1A 0A. The format is lossless and can carry an alpha channel. image-convert matches all eight bytes before sharp is loaded, while the OCR detector settles for the first four. Both services then read the dimensions from the header without decoding the raster, and a file whose width times height exceeds the shared decoded-pixel budget is refused at that point. The check exists because a small, highly compressible PNG, such as a solid colour or a decompression bomb, can inflate to an enormous raster, and Tesseract's cost grows with the pixel count. Since PNG stores exact pixel values, a PNG source contributes no compression noise of its own; any loss in a converted file comes from the lossy target encoder.

plain text as a target

Plain text is recovered in several ways, each tied to one reader. ocr-image-to-text reads pixels with Tesseract running offline from local files, returns a mean confidence with the text, and offers only the languages whose data ships with the service. pdf-to-text returns the existing text layer page by page and performs no OCR. epub-to-text follows the spine's reading order and adds title, creator and language metadata. pptx-to-text returns slide body text in slide order, leaving out speaker notes. html-clean-text extracts article text with Readability by default and returns Markdown in the same response. Because each reader works differently, the text reflects its source: recognised characters with a confidence score from an image, but characters already present in the file from documents and pages.

04Errors

Code	HTTP
`invalid_input`	`400`
`blocked_target`	`403`
`unreachable_target`	`424`
`upstream_timeout`	`424`
`unsupported_content_type`	`415`
`too_large`	`413`
`unprocessable`	`422`
`internal`	`500`

When each is raised, and what it means for payment: /docs/ocr-image-to-text.

05Questions

What does it cost to convert PNG to plain text?: $0.005 per call, paid in USDC on Base via x402.
How large can the PNG file be?: Up to 10 MB (10,485,760 bytes). Past a limit the call answers too_large (413).
What happens if the conversion fails?: The tool answers with a typed JSON error from the errors table. Payment is settled only after the tool has produced its result; a call that fails inside the tool is never settled.