Real-ESRGAN at a glance
Checked 16 Oct 2026Real-ESRGAN is the model family most free upscalers build on (Upscayl, Video2X, chaiNNer). It offers 4x general and anime models, a Python script with optional GFPGAN face enhancement and tiling for low VRAM, and portable realesrgan-ncnn-vulkan binaries for Windows, Linux and macOS that need no CUDA. Code and released weights are BSD-3-Clause. It is a command-line tool with no GUI. Development has stopped: last commit Aug 2024, last release Sep 2022, binary Apr 2022.
RepoQuarry. “Real-ESRGAN.” Checked 16 Oct 2026. https://repoquarry.dev/projects/real-esrganHow much of Topaz Gigapixel does it replace?
4 of 6 jobs. Command-line 4x upscaling of files or folders, anime models, and optional GFPGAN face enhancement; no GUI.
For recover and redefine low-quality or compressed images, Photoshop and Lightroom plugin, see other Topaz Gigapixel alternatives. You may also need to build an interface, an API server and hosting yourself.
Facts, with sources
Every answer links to where we found it. We read the documentation; we did not install it.
This executable file is portable and includes all the binaries and models required. No CUDA or PyTorch environment is needed.github.com · Real-ESRGAN#portable-executable-files-ncnn ↗ today
--face_enhance Whether to use GFPGAN to enhance face. Default: Falsegithub.com · Real-ESRGAN#python-script ↗ today
executable files for Intel/AMD/Nvidia GPUgithub.com · Real-ESRGAN#portable-executable-files-ncnn ↗ today
You can download Windows / Linux / MacOS executable files for Intel/AMD/Nvidia GPU.github.com · Real-ESRGAN#portable-executable-files-ncnn ↗ today
./realesrgan-ncnn-vulkan.exe -i input.jpg -o output.png -n model_namegithub.com · Real-ESRGAN#portable-executable-files-ncnn ↗ today