AI Image Upscaler: 4x Real-ESRGAN, No GPU

Make small, blurry or compressed images 2x or 4x bigger and sharper with Real-ESRGAN, one of the best-known open-source AI super-resolution models. Upscale product photos, old pictures, illustrations, thumbnails, screenshots and AI-generated images in bulk. Just paste image links.

Pay per image. No subscription, no GPU, no software to install. A cheap alternative to Topaz Gigapixel, Let's Enhance or Upscayl in the cloud, and easy to call from Make, Zapier, n8n or your own code.

Before and after

A real 233×263 px product photo, upscaled 4x by this Actor in 1.6 seconds (left: the same photo simply enlarged):

Zoomed detail of a 250×300 px photo: fur, whiskers and eyes are restored instead of blocky pixels.

Features

• 4x or 2x AI upscaling (the 2x mode runs the 4x model on a half-size copy, which looks sharper than simply resizing).
• Removes JPEG artefacts and noise while restoring edges and texture.
• Bulk: any number of image URLs in one run.
• PNG, JPEG or WebP output. Transparency is kept for PNG and WebP.
• Phone photos are rotated correctly (EXIF orientation).
• Size guard: the output is capped at a megapixel limit you choose (default 16 MP), and bigger inputs are reduced first.

Output

Each upscaled image is saved to the run's key-value store, and the dataset gets a direct link:

{
  "sourceUrl": "https://example.com/product-small.jpg",
  "imageUrl": "https://api.apify.com/v2/key-value-stores/<id>/records/upscaled-0001.png",
  "scale": 4,
  "originalWidth": 320,
  "originalHeight": 213,
  "width": 1280,
  "height": 852,
  "megapixels": 1.09,
  "format": "png",
  "processingMs": 4200,
  "billedImages": 1
}

Pricing

You pay per upscaled image for outputs up to 4 megapixels (e.g. 2048×2048). Bigger outputs count as one image per started 4 MP. Failed downloads and unsupported files are never charged.

Use cases

• E-commerce: enlarge supplier product photos to meet marketplace size requirements.
• Print: upscale images for posters, flyers and merch.
• AI art: upscale Midjourney, Stable Diffusion or DALL·E outputs.
• Archives: improve old, low-resolution photos and scans.
• Automation: add an upscaling step to your image pipeline via API.

Notes

• The AI adds plausible detail. It cannot recover information that is not in the image (e.g. unreadable text).
• Faces are upscaled like the rest of the image. There is no separate face-restoration model.
• Model: Real-ESRGAN realesr-general-x4v3 by Xintao Wang et al., BSD-3-Clause license.
• Only upload images you have the rights to use.

Questions or ideas?

Open an issue on the Issues tab. Feature requests (background removal, face restoration…) are welcome.

Chain it after any product scraper

Already scraping products on Apify (Shopify, Amazon, WooCommerce, Google Shopping, marketplaces…)? Upscale the product images straight from the scraper's results, with no copying of URLs:

1. Open your scraper's Integrations tab and add Run Actor, then choose this Actor.
2. Use this input:

{
  "datasetId": "{{resource.defaultDatasetId}}",
  "imageUrlField": "images",
  "maxImagesPerItem": 1
}

3. Every result row keeps itemUrl, itemId, itemTitle and datasetItemIndex from the source item, so you can join the new images back to your products.

You can also pick any existing dataset in the Images from a dataset field in the Console. Leave Image field in dataset empty and common fields (imageUrl, image, images, thumbnail, …) are detected automatically. Dot paths such as product.images.0.src and protocol-relative Shopify CDN links (//cdn.shopify.com/...) work too.

How to use it via API

You can run the Actor from the Apify Console, on a schedule, or from your own code. Get your API token in Apify Console → Settings → Integrations.

Python

from apify_client import ApifyClient

client = ApifyClient("<YOUR_API_TOKEN>")
run = client.actor("digitalni.produkty.pro.zivot/ai-image-upscaler").call(run_input={
    "imageUrls": [
        "https://upload.wikimedia.org/wikipedia/commons/thumb/a/a7/Camponotus_flavomarginatus_ant.jpg/330px-Camponotus_flavomarginatus_ant.jpg"
    ],
    "scale": "4"
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item)

JavaScript / Node.js

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_API_TOKEN>' });
const run = await client.actor('digitalni.produkty.pro.zivot/ai-image-upscaler').call({
    "imageUrls": [
        "https://upload.wikimedia.org/wikipedia/commons/thumb/a/a7/Camponotus_flavomarginatus_ant.jpg/330px-Camponotus_flavomarginatus_ant.jpg"
    ],
    "scale": "4"
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

Integrations and AI agents

• Export results as JSON, CSV, Excel, XML or HTML, or open them directly in Google Sheets.
• Connect to Make, Zapier, n8n, Slack, Google Drive, Airbyte or any webhook to get AI-upscaled images into your workflow automatically.
• Use it from AI agents and LLM apps (Claude, ChatGPT, Cursor, LangChain…) through the Apify MCP server: the agent can call this Actor as a tool.
• Schedule runs (hourly, daily, weekly) to keep data fresh without any code.

FAQ

How much does it cost? $0.01 per output image up to 4 megapixels (larger outputs count per started 4 MP). 1,000 product photos cost about $10.

Do I need a GPU or software? No. Real-ESRGAN runs in the cloud on CPU. Just provide image URLs.

Are my images stored? Only in your own Apify storage for the run, under your account's data retention settings.

More tools from the same developer

All tools with code examples: github.com/Phenixik/apify-actors

• Sitemap URL Extractor: every URL from a website's sitemaps
• Broken Link Checker: find 404s and dead links on your site
• PageSpeed & Lighthouse Audit: bulk Core Web Vitals and SEO scores
• Website Screenshot: full-page, mobile and PDF screenshots
• Domain WHOIS & DNS Lookup: expiry, DNS, SPF/DMARC and SSL for many domains
• PDF Text Extractor: text, Markdown and RAG chunks from PDFs
• RSS Feed Reader & Monitor: only-new-items feed monitoring
• ATS Jobs: jobs from Greenhouse, Lever, Ashby & more
• App Store Reviews: iOS reviews from 50+ countries