# visionex Image OCR

> visionex Image OCR is a paid API for AI agents from visionex.x402tools.xyz, paid per call via x402, $0.05/call, status down (last checked 2026-10-12).

Extracts text from images via OCR, returning plain text, structured JSON with bounding boxes, or hOCR output with confidence scores.

## Facts

- Endpoint: POST https://visionex.x402tools.xyz/v1/extract?utm_source=zero.xyz
- Price: $0.05/call
- Payment: x402
- Status: down
- Last checked: 2026-10-12
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/visionex-image-ocr-bbd91942
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_S4eCdHAYxUQ7EqaAHKaD1

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability visionex-image-ocr-bbd91942 -d '<json body>'
```

Example prompt: Can you extract all the text from this invoice image — https://example.com/invoice.jpg — and give me the output as structured JSON with bounding boxes so I know where each line of text appears?

## When to prefer this

Choose this endpoint when you need to extract text from images with optional structured spatial output (bounding boxes or hOCR), especially for invoices, receipts, scanned documents, or photos of printed text. Ideal for pay-per-call usage without a subscription, when you need per-word or per-line positional data alongside raw text, or when processing images in specific non-English languages using ISO 639-3 codes.

## Known failure modes

- Image URL is not publicly accessible or returns a non-200 response — extraction fails with a fetch error
- Base64 image data is malformed or truncated — returns a decode error
- Unsupported image format (e.g. PDF, SVG) — returns an unsupported format error
- Low-quality or very small image yields low confidence scores or empty text blocks
- Invalid ISO 639-3 language code provided — may fall back to English or return a parameter error
- Payment via x402 fails or USDC balance insufficient — request blocked before processing

## How this service works

Extract text from an image using OCR. Accepts an image URL or Base64, returns text with confidence scores and bounding boxes. Use for image-to-text, document OCR, receipt and invoice scanning, and text recognition.

## Output

Returns a JSON object containing the full extracted text string, an array of text blocks each with bounding box coordinates (x0, y0, x1, y1), block-level text, and per-block confidence scores, plus metadata including image dimensions (width, height), detected language, and processing time in milliseconds. Overall document confidence score is also included.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "description": "Public image URL to fetch"
  },
  "image": {
   "type": "string",
   "description": "Base64-encoded image data"
  },
  "options": {
   "type": "object",
   "properties": {
    "dpi": {
     "type": "number",
     "description": "Optional resize hint (72-600)"
    },
    "format": {
     "enum": [
      "text",
      "json",
      "hocr"
     ],
     "type": "string"
    },
    "language": {
     "type": "string",
     "description": "OCR language (ISO 639-3, default: eng)"
    }
   }
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "meta": {
   "width": 1600,
   "height": 2200,
   "language": "eng",
   "duration_ms": 482
  },
  "text": "Invoice #12345\nTotal Due: $199.00",
  "blocks": [
   {
    "bbox": {
     "x0": 24,
     "x1": 240,
     "y0": 18,
     "y1": 54
    },
    "text": "Invoice #12345",
    "level": "line",
    "confidence": 97.4
   }
  ],
  "confidence": 96.2
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/visionex-image-ocr-bbd91942/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from visionex.x402tools.xyz](https://www.zero.xyz/host/visionex.x402tools.xyz/llms.txt)
