# Image & Scanned PDF to Text (OCR)

> Image & Scanned PDF to Text (OCR) is a paid MCP server for files and documents. Apify-hosted MCP server for OCR on images and scanned PDFs in 41 languages. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Paid
- **Platforms:** API, Web
- **Website:** https://apify.com/prometheus-maker/image-to-text-ocr
- **Repository:** https://github.com/prometheus-maker/apify-tools-examples
- **MCP Registry name:** io.github.prometheus-maker/image-ocr
- **Categories:** files and documents, design and media
- **Listing updated:** 2026-11-14

## About

The server runs two OCR engines and picks one per file. A photo engine (PaddleOCR models via RapidOCR) handles signs, menus, receipts and screenshots; the page claims it found 89% of words on 80 receipt photos versus 38% for Tesseract on its own test. A document engine (Tesseract 5) covers scans and PDFs in all 41 languages, including Cyrillic, Greek, Arabic, Hebrew, Hindi, Thai and Korean. Auto mode chooses between them and each result says which read it.

Each image or PDF page returns paragraph-grouped text, an average confidence from 0 to 100, a word count and the image's sha256. Optional line output adds each line's confidence and a bounding box as fractions of image size, handy for highlighting or redaction. Lines below a confidence threshold (default 30) are dropped as noise.

Formats include JPG, PNG, WebP, TIFF, GIF, BMP and PDF, with the first 50 pages of each PDF read by default. It starts at $3.00 per 1,000 reads.

## Features

- **Two OCR engines:** Photo and document engines chosen automatically, with the engine named in each result.
- **41 languages:** Covers Latin, Cyrillic, Greek, Arabic, Hebrew, Indic, Thai, Korean, Japanese and Chinese scripts.
- **Confidence scores:** Average confidence per result for flagging poor scans.
- **Line bounding boxes:** Optional line text with positions for highlighting or redaction.
- **Scanned PDF support:** Each PDF page is returned as its own result with page and pageCount.
- **Layout modes:** Auto, sparse for scattered text, and single-line for captions.

## Good for

- Extract text from receipt photos
- Make scanned PDFs searchable
- Pull text from screenshots for an LLM
- Highlight or redact text using line positions

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http image-ocr 'https://mcp.apify.com/?tools=prometheus-maker/image-to-text-ocr'
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://mcp.apify.com/?tools=prometheus-maker/image-to-text-ocr
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "image-ocr": {
      "url": "https://mcp.apify.com/?tools=prometheus-maker/image-to-text-ocr"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "image-ocr": {
      "type": "http",
      "url": "https://mcp.apify.com/?tools=prometheus-maker/image-to-text-ocr"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "image-ocr": {
      "serverUrl": "https://mcp.apify.com/?tools=prometheus-maker/image-to-text-ocr"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "image-ocr": {
      "type": "http",
      "url": "https://mcp.apify.com/?tools=prometheus-maker/image-to-text-ocr"
    }
  }
}
```

## FAQ

### What does it cost?

From $3.00 per 1,000 image or page reads, as listed on the Apify page.

### Which files can it read?

JPG, PNG, WebP, TIFF, GIF, BMP and PDF.

### How many PDF pages are read?

The first 50 pages of each PDF by default; the limit is adjustable.

### Can I tell how reliable a read is?

Yes, each result has an average confidence from 0 to 100.

### Is Image & Scanned PDF to Text (OCR) free?

No. Image & Scanned PDF to Text (OCR) is a paid product.

### How do I add Image & Scanned PDF to Text (OCR) to Claude Code?

Run this command in your terminal: claude mcp add --transport http image-ocr 'https://mcp.apify.com/?tools=prometheus-maker/image-to-text-ocr'

### What are the best alternatives to Image & Scanned PDF to Text (OCR)?

The closest alternatives to Image & Scanned PDF to Text (OCR) are HyperFrames by HeyGen, ffmpeg-skill, ASTra, AvatarSay — Fictional Character & Talking Video MCP and ChartAnt.

## Alternatives

- [HyperFrames by HeyGen](https://ai-powered.com/mcp/hyperframes-by-heygen): HyperFrames by HeyGen is a freemium MCP server for design and media.
- [ffmpeg-skill](https://ai-powered.com/mcp/ffmpeg-skill): ffmpeg-skill is an open-source MCP server for design and media.
- [ASTra](https://ai-powered.com/mcp/astra): ASTra is a freemium MCP server for files and documents.
- [AvatarSay — Fictional Character & Talking Video MCP](https://ai-powered.com/mcp/avatarsay-fictional-character-and-talking-video-mcp): AvatarSay — Fictional Character & Talking Video MCP is a freemium MCP server for design and media.
- [ChartAnt](https://ai-powered.com/mcp/chartant): ChartAnt is a freemium MCP server for design and media.
- [VfxSmith](https://ai-powered.com/mcp/vfxsmith): VfxSmith is a freemium MCP server for design and media.
- [CalderFlow](https://ai-powered.com/mcp/calderflow): CalderFlow is a free MCP server for design and media.
- [Hourlore](https://ai-powered.com/mcp/hourlore): Hourlore is a freemium MCP server for design and media.

---
Source: https://ai-powered.com/mcp/image-and-scanned-pdf-to-text-ocr
