# AI use cases for Upscayl (upscayl/upscayl) Repo: [upscayl/upscayl](https://github.com/upscayl/upscayl) | Use cases: 3 | Page: https://dedupedai.com/use-cases/upscayl-upscayl/ Upscayl is a free and open-source AI image upscaler for Linux, macOS and Windows. It enlarges and enhances low-resolution and pixelated images with Real-ESRGAN models on a Vulkan-compatible GPU, supports batch upscaling and custom models, and has a separate CLI called upscayl-ncnn. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Upscale old or low-resolution photos locally on your Mac: https://dedupedai.com/use-cases/upscayl-upscayl/upscale-old-photos-on-your-mac-with-upscayl/ - Batch upscale a whole folder of images on Windows or Linux: https://dedupedai.com/use-cases/upscayl-upscayl/batch-upscale-a-folder-of-images-with-upscayl/ - Add custom models or run Upscayl from source: https://dedupedai.com/use-cases/upscayl-upscayl/add-custom-upscaling-models-or-build-upscayl-from-source/ ## How to upscale old or low-resolution photos locally on your Mac with Upscayl Repo: [upscayl/upscayl](https://github.com/upscayl/upscayl) | Difficulty: Beginner | Time: 10 minutes | Basis: From the README Install Upscayl with Homebrew and enlarge images on your own GPU, free. **Who it is for:** Someone with old family photos or small web images who needs them larger for print or a site. **The problem:** Online upscalers charge per image or upload your photos to someone else's server. **The outcome:** Larger, sharper versions of your images made locally. ### Steps 1. Run `brew install --cask upscayl` (macOS 12 or later) 2. Open Upscayl from Applications; the first time, right click it and choose Open 3. Choose an image, pick a model such as High Fidelity or Ultrasharp, and an output folder 4. Upscale and compare before and after 5. If a photo is out of focus, expect little change: Upscayl cannot de-blur ### Prompt to paste into your agent ```text Install upscayl/upscayl with Homebrew, then upscale the images in and save them to . Tell me which model you used and flag any that look unchanged. ``` ### Caveat This card is written from the upscayl/upscayl README, not from a test run by Deduped. Before you start, make sure you have a Vulkan-compatible GPU. Check the repo for newer instructions before you rely on these steps. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/upscayl-upscayl/upscale-old-photos-on-your-mac-with-upscayl/ ## How to batch upscale a whole folder of images on Windows or Linux with Upscayl Repo: [upscayl/upscayl](https://github.com/upscayl/upscayl) | Difficulty: Beginner | Time: 20 minutes | Basis: From the README Install Upscayl on Windows or Linux and run Batch Upscayl on a folder. **Who it is for:** A small shop owner or game modder with dozens of small images to enlarge at once. **The problem:** Upscaling images one by one takes too long. **The outcome:** A folder of upscaled images, post-processed after the batch finishes. ### Steps 1. On Windows, download upscayl-x.x.x-win.exe from the releases section and run it 2. On Linux, download the AppImage, allow it to execute, and run it (RPM, DEB and Flatpak also exist) 3. Turn on batch mode and choose your input folder 4. Set the GPU ID if you have more than one GPU 5. Let the batch finish completely before checking results, as the FAQ explains ### Prompt to paste into your agent ```text Install upscayl/upscayl on my machine and batch upscale every image in . Do not interrupt the batch, then check that all outputs have the expected scale. ``` ### Caveat This card is written from the upscayl/upscayl README, not from a test run by Deduped. Before you start, make sure you have a Vulkan-compatible GPU. Check the repo for newer instructions before you rely on these steps. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/upscayl-upscayl/batch-upscale-a-folder-of-images-with-upscayl/ ## How to add custom models or run Upscayl from source with Upscayl Repo: [upscayl/upscayl](https://github.com/upscayl/upscayl) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Load extra models from upscayl/custom-models, or run the app from source with npm. **Who it is for:** A tinkerer who wants anime or texture specific models, or to change the app. **The problem:** The default models are general purpose and may not suit your images. **The outcome:** Upscayl running with custom models, or a local dev build you can change. ### Steps 1. Get more models from github.com/upscayl/custom-models and point Upscayl at their folder 2. To run from source, `volta install node`, then `git clone https://github.com/upscayl/upscayl` 3. cd Upscayl, run `npm install`, then `npm run start` 4. Package your own build with `npm run dist` 5. For scripts, use the separate upscayl-ncnn CLI ### Prompt to paste into your agent ```text Clone upscayl/upscayl, run it from source with `npm run start`, add the from upscayl/custom-models, and compare it with the default model on . ``` ### Caveat This card is written from the upscayl/upscayl README, not from a test run by Deduped. Before you start, make sure you have a Vulkan-compatible GPU. Check the repo for newer instructions before you rely on these steps. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/upscayl-upscayl/add-custom-upscaling-models-or-build-upscayl-from-source/