Screenshots Options

Screenshots options allow you to configure whether page screenshots are generated during document parsing.

How It Works

When enabled, this feature takes a screenshot of each page in the document, providing a visual representation of the original page layout alongside the extracted text content.

Configuration

In v2, screenshot generation is controlled via the images_to_save array. Include "screenshot" in the array to enable screenshot generation.

Enable Screenshots

Enable screenshot generation for each page of the document:

{
  "output_options": {
    "images_to_save": ["screenshot"]
  }
}

Combine with Other Image Types

You can combine screenshots with other image types:

{
  "output_options": {
    "images_to_save": ["screenshot", "embedded", "layout"]
  }
}

Disable Screenshots

To disable screenshots, simply omit "screenshot" from the array, or use an empty array to disable all image saving:

{
  "output_options": {
    "images_to_save": []
  }
}

Note: If images_to_save is not specified, no images (including screenshots) are saved by default.

Complete API Request Example

cURL

Parse the Document with Screenshot Generation Enabled

curl -X 'POST' \
  'https://api.cloud.llamaindex.ai/api/v2/parse' \
  -H 'Accept: application/json' \
  -H 'Content-Type: application/json' \
  -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY" \
  --data '{
    "file_id": "<file_id>",
    "tier": "cost_effective",
    "version": "latest",
    "output_options": {
      "images_to_save": ["screenshot"]
    }
  }'

Retrieving Screenshots

After parsing completes, retrieve screenshot metadata using the images_content_metadata expand parameter. This returns presigned URLs for direct download:

curl -X 'GET' \
  'https://api.cloud.llamaindex.ai/api/v2/parse/{job_id}?expand=images_content_metadata' \
  -H 'Accept: application/json' \
  -H "Authorization: Bearer $LLAMA_CLOUD_API_KEY"

The response includes metadata with presigned URLs:

{
  "job": { ... },
  "images_content_metadata": {
    "total_count": 5,
    "images": [
      {
        "filename": "screenshot_page_0.png",
        "content_type": "image/png",
        "size_bytes": 45678,
        "presigned_url": "https://s3.amazonaws.com/..."
      },
      {
        "filename": "screenshot_page_1.png",
        "content_type": "image/png",
        "size_bytes": 52341,
        "presigned_url": "https://s3.amazonaws.com/..."
      }
    ]
  }
}

Use the presigned_url for each image to download the screenshots directly. URLs are temporary and valid for a limited time.

Python

import httpx
from llama_cloud import LlamaCloud


client = LlamaCloud(api_key="LLAMA_CLOUD_API_KEY" )

# Upload and parse a document, requdesting image content metadata
result = await client.parsing.parse(
    upload_file="example_file.pdf",
    tier="agentic",
    version="latest",
    output_options={
        "screenshots": {"enable": True},
    },
    expand=["images_content_metadata"],
)

# Download extracted images using presigned URLs
for image in result.images_content_metadata.images:
    print(f"Downloading {image.filename}, {image.size_bytes} bytes")
    with open(f"{image.filename}", "wb") as img_file:
        with httpx.Client() as http_client:
            response = http_client.get(image.presigned_url)
            img_file.write(response.content)

TypeScript

import fs from "fs";
import { LlamaCloud } from "@llamaindex/llama-cloud";

const client = new LlamaCloud({
  apiKey: "LLAMA_CLOUD_API_KEY",
});

const result = await client.parsing.parse({
  upload_file: fs.createReadStream('example_file.pdf'),
  tier: "agentic",
  version: "latest",
  expand: ["images_content_metadata"],
  output_options: {
    screenshots: {
      enable: true
    }
  }
});

for (const image of result.images_content_metadata!.images) {
  console.log(`Downloading ${image.filename}, ${image.size_bytes} bytes`);
  const response = await fetch(image.presigned_url);
  const arrayBuffer = await response.arrayBuffer();
  fs.writeFileSync(image.filename, Buffer.from(arrayBuffer));
}