Topaz Labs makes photo and video enhancement software based on deep learning, a kind of machine learning that uses neural networks with many layers. Its models upscale images and video, which means they add pixels to raise the resolution, and they also reduce noise, sharpen and restore detail.
The models run locally in desktop apps and in the cloud. Topaz Photo and Topaz Gigapixel work on images; Gigapixel upscales an image up to 16 times its pixel count. The web app for video offers precision and creative upscaling, frame interpolation, slow motion and SDR to HDR conversion. The Topaz API gives other products the same models through a REST API, an interface that a program calls over HTTP: image upscaling, denoise and sharpen, video upscaling and video frame interpolation, with upscaling models made for AI-generated images and video.
The apps are sold as subscriptions, and the Topaz Studio plan includes all of them. The API is a monthly plan with a set number of credits; one credit processes one image request of up to 24 megapixels of output, and unused credits roll over.