upscayl/upscayl

🆙 Upscayl - #1 Free and Open Source AI Image Upscaler for Linux, MacOS and Windows.

What it solves

Upscayl is a free and open-source AI image upscaler that allows users to enlarge and enhance low-resolution or pixelated images without losing quality.

How it works

It utilizes AI models, specifically Real-ESRGAN, and the Vulkan architecture to guess missing details and enhance image resolution. The application requires a Vulkan-compatible GPU to function.

Who it’s for

Anyone looking to increase the resolution of their images locally on Windows, macOS, or Linux, provided they have a compatible graphics card.

Highlights

  • Cross-platform support for Windows, macOS, and Linux.
  • Based on the Real-ESRGAN AI model.
  • Supports batch upscaling of images.
  • Ability to use custom AI models.
  • Fully open-source backend (AGPLv3 license).

Related

  • Project
    YukihoAA/waifu2x_snowshell

    A GUI shell for various AI image upscaling tools like waifu2x and Real-ESRGAN, enabling easy image enhancement through a drag-and-drop interface.

  • Project
    EutropicAI/Final2x

    Final2x is a cross‑platform desktop app for AI‑based image super‑resolution. It bundles ready‑made models, supports custom PyTorch models, and runs on Windows, macOS, and Linux via an Electron‑Vite UI backed by the open‑source `cccv` restoration library.

  • Project
    thekevinscott/UpscalerJS

    A JavaScript library for AI-powered image enhancement, allowing developers to increase resolution, denoise, and deblur images in the browser or Node.js.

  • Project
    nihui/zimage-ncnn-vulkan

    A portable ncnn-based implementation of the Z-Image generator that enables high-performance image generation, inpainting, and ControlNet support on any Vulkan-capable GPU.

  • Project
    the-database/mpv-AnimeJaNai

    A custom mpv video player build and a set of Real-ESRGAN models that enable real-time 4K upscaling of anime videos using TensorRT or DirectML.