Real-ESRGAN

Real-ESRGAN develops algorithms for general image and video restoration.

github.com/xinntao/Real-ESRGAN
Real-ESRGAN homepage screenshot

About Real-ESRGAN

Real-ESRGAN is an open-source project for general image and video restoration and upscaling. It extends ESRGAN and is trained with synthetic data, with Python scripts for image and video inference and separate models for general scenes and anime images and videos. It supports tiled inference, images with alpha channel, gray and 16-bit images, arbitrary output scaling, finetuning on own or paired data, and face enhancement through GFPGAN integration. Local use requires Python >= 3.7 and PyTorch >= 1.7 with manual installation, while portable executables and listed online demos provide other ways to try it.

Key features

  • Restore and upscale general images and videos
  • Anime-specific models for images and videos
  • Face enhancement through GFPGAN integration
  • Tiled inference with support for alpha, gray, 16-bit images
  • Arbitrary output scaling with LANCZOS4 resizing
  • Finetuning support on own or paired data
Best for: People with a Python and PyTorch setup who want to run image and video restoration locally, including anime content.
Good for
  • Open-source repository with training code released
  • Includes models for general scenes and anime images and videos
  • Offers inference by Python script, portable executables, and online demos
Watch out for
  • Requires Python >= 3.7 and PyTorch >= 1.7 with manual installation steps
  • ARC online demo only supports the RealESRGAN_x4plus_anime_6B model
  • Local use involves cloning the repository and installing multiple packages

Real-ESRGAN at a glance

Type
Open-source project
Pricing
Not confirmed. Check the site
Open source
Yes · Source code
Language
English
Last checked
Oct 5, 2026 · Online when checked
Listed
Oct 5, 2026

Real-ESRGAN alternatives

Tools and sites you could use instead, ranked by how closely they match.

All 24 alternatives

Face Restoration

github.com/TencentARC/GFPGAN

GFPGAN develops algorithms for restoring real-world faces in images, with its repository focused on practical face-restoration research.

Open source
18 alternatives

Popular in Image Creation & Editing

Well-known listings from the same topic.

All 244

Real-ESRGAN FAQ

What is needed to install Real-ESRGAN?
Python >= 3.7, with Anaconda or Miniconda recommended, and PyTorch >= 1.7, plus packages including basicsr, facexlib, gfpgan, and requirements.txt.
What are the ways to run inference?
Online inference, portable executable files using NCNN, and Python scripts.
Does it support face enhancement?
Yes, it integrates GFPGAN to support face enhancement.
Is something wrong or out of date?

Tell us and an editor will check it.

Own this website?

Claim the listing and suggest updates yourself. An editor checks every claim first.

Claim this listing