flowframes

Why we included this project

Video editors who work with low-frame-rate footage will find a practical local solution here. This Windows application bundles several AI interpolation models, including RIFE, DAIN, and FLAVR, behind a point-and-click interface, letting you convert 24 or 30 FPS clips to 60, 120, or custom rates without writing code or managing Python environments. The NCNN backend runs on any Vulkan-capable GPU, while the PyTorch paths get more out of Nvidia cards, and you can keep audio, subtitles, and metadata from the source. Since everything runs on your own hardware, it's a good fit for anyone who wants reliable frame interpolation without sending footage to a cloud service. The project is donationware, and the code is complete enough to compile yourself if you prefer not to use the release builds.