Free and open-source AI image upscaler API powered by Real-ESRGAN. Upscale images 2x, 3x, or 4x with privacy-first local processing, batch support, and no watermarks. Self-host or integrate easily with REST endpoints. Designed for developers, creators, and researchers!
| Original | Enhanced 4x |
|---|---|
 |
 |
Upscale your images by 4x, preserve details, remove artifacts, and enhance visual quality. See the difference instantly!
- 🎯 4x Image Upscaling – High-fidelity enhancement via Real-ESRGAN
- 🆓 100% Free & Open Source – MIT license, no hidden costs
- 🏠 Self-Hosted / Local Processing – No cloud uploads, maximize privacy
- 🔌 REST API & Web UI – Flexible endpoints and user-friendly browser app
- 🚫 No Watermarks/Limitations – Unlimited usage for all users
- 🍎 Cross-Platform – Run on Windows, Linux, macOS, Apple Silicon
- 📦 Batch Support – Upscale multiple images at once
- ⚡ Fast Results – Leverages Vulkan for speed, no CUDA required
- 📝 Comprehensive Docs – Easy integration with other apps, sites, or workflows
git clone https://github.com/kalprajsolutions/image-enhance-api.git
cd image-enhance-api
npm installPlace your realesrgan-ncnn-vulkan executable in the project root (Windows version included).
npm startOpen http://localhost:5000 in your browser to use the web interface for uploading and enhancing images.
Lists available enhancement models.
[
"realesr-animevideov3-x2",
"realesr-animevideov3-x3",
"realesr-animevideov3-x4",
"realesrgan-x4plus",
"realesrgan-x4plus-anime"
]Enhance a single image.
Form Fields:
image(file): Image to enhancemodel(optional): Model to use, defaults torealesrgan-x4plus
Example Request:
curl -X POST -F "image=@/path/to/your/image.jpg" -F "model=realesrgan-x4plus-anime" http://localhost:5000/enhanceSuccess Response:
{
"success": true,
"file": "http://localhost:5000/output/xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx.png"
}- realesr-animevideov3-x2
- realesr-animevideov3-x3
- realesr-animevideov3-x4
- realesrgan-x4plus (default)
- realesrgan-x4plus-anime
Supports upscaling of photography, anime, art, and more.
- Photography Restoration – Sharpen and restore low-res images
- Content Creation – Enhance digital art, anime, and game assets
- Web & App Integration – Seamlessly add enhancement via your own API calls
- Batch Processing – Optimize large image sets for e-commerce, printing, or archives
- Research & AI – Use for datasets, training, enhancement workflows
Extract, enhance, and merge video frames using FFmpeg and Real-ESRGAN:
ffmpeg -i input.mp4 tmp_frames/frame%08d.jpg
./realesrgan-ncnn-vulkan.exe -i tmp_frames -o out_frames -n realesr-animevideov3-x2 -f jpg
ffmpeg -i out_frames/frame%08d.jpg -i input.mp4 -map 0:v:0 -map 1:a:0 -c:a copy -c:v libx264 -r 23.98 -pix_fmt yuv420p enhanced_video.mp4Supports Docker, local Node.js, and direct CLI usage. Easily integrate into CI/CD or cloud workflows.
- Completely Free & Open Source – Outperforms paid SaaS upscalers
- Privacy & Ownership – No third-party uploads; process images on your hardware
- Easy Integration – Simple REST endpoints, web UI, and batch support
- Latest Real-ESRGAN Models – Cutting-edge fidelity, anime and photographic models
- Community Driven – Accepting contributions and feature requests
This project is licensed under the MIT License. See LICENSE.
- Built on Real-ESRGAN by Xintao Wang et al.
- Executable powered by realsr-ncnn-vulkan by nihui
- Tencent/ncnn
- Open source by kalprajsolutions.com
- Report Bugs
- Request Features
- Discussions
- Star & share to support free AI!
real-esrgan, image upscaler, ai image enhancement, free api, open source, super resolution, anime upscaler, photo enhancement, rest api, batch processing, privacy, local processing, cross-platform, node.js
Made with ❤️ for the open source community