Double the resolution of a photo, drawing or screenshot with AI. It runs on your own GPU, so the image is never uploaded and there is no watermark.
Drag the slider on the result to compare the original with the upscaled image.
Resizing in an image editor spreads each pixel over a bigger area, which makes edges soft and blocky. An AI upscaler uses a neural network trained on low- and high-resolution pairs, so it redraws edges and lines sharply at the new size.
The networks here come from the open-source WebSR project (based on Anime4K). They are fast and clean, and they work best on line art, anime, logos and screenshots. Photos get sharper edges, but missing detail such as skin texture is not invented.
Yes, with no signup, watermark or daily limit. The work happens on your own device.
No. The image is processed with WebGPU inside your browser and never sent to a server.
Each run doubles the size. For 4x, download the result and upscale it again (as long as it stays under 4096 pixels per side).
Yes, use the free AI video upscaler. It upscales every frame and keeps the original audio.