Concat is a free, open-source CapCut alternative that runs its AI locally
Concat is an open-source video editor for macOS, Windows, Linux and Android with auto-captions, text-to-speech with voice cloning and background removal that run on your own machine. It's in beta, has no watermark or account, and scripts or AI agents can drive it through a JSON-RPC API.
What it does
It covers multi-track cutting with keyframes and a curve editor, about 100 effects, filters and transitions, local Whisper auto-captions, text-to-speech and voice cloning from a few seconds of a recording, background removal, voice cleanup, and export in H.264, HEVC or AV1 up to 4K 60 with 10-bit colour. It runs on a native Rust engine with a GPU compositor.
The AI models download once from Settings and then run offline, and the developers say your footage never leaves your disk.
Results and limits
Version 0.2.7 (13 October) fixes exports that lost the GPU on Vulkan and adds caption generation from a clip's right-click menu. Its notes were corrected on 16 October: a blur or glow can still stall an export on Windows (issue #297).
Worth knowing
The minimum is a 64-bit CPU and 4 GB of RAM, with no GPU required; the recommended setup is 6+ cores, a GPU with Metal, DirectX 12 or Vulkan, and 16 GB of RAM. All the optional models together take about 3.4 GB. It's licensed AGPL-3.0, and on iOS it has to be sideloaded.
Try it yourself
Import [a 60-second talking-head clip], right-click it and generate captions, turn on voice cleanup, then export a 9:16 version at 1080p.