Guides
System requirements
Concat runs everything on your machine, so the hardware sets the ceiling.
Minimum is what a build runs on at all. Recommended is what makes 1080p editing feel smooth and keeps 4K exports and captions from being a wait.
| Minimum | Recommended | |
|---|---|---|
| CPU | Any 64-bit processor from 2013 or later | 6 cores or more |
| GPU | None. Without a usable GPU the window and monitor fall back to the CPU | Any GPU with Metal (macOS), DirectX 12 (Windows) or Vulkan (Linux) |
| RAM | 4 GB | 16 GB for 4K timelines and the larger caption models |
| Storage | 500 MB for the app and the smallest caption model | About 3.4 GB for every optional model, plus room for projects and exports |
Optional models
Auto-captions, voices, cutout and enhance use models that download from Settings on first use and then never need the network again. Sizes by family:
| Family | Models | Size |
|---|---|---|
| Auto-captions (Whisper) | Tiny, Base or Small, each in English or multilingual | 78 MB, 148 MB or 488 MB each |
| Text-to-speech (Kokoro) | Built-in speakers | 132 MB (int8) or 350 MB |
| Voice cloning (Pocket TTS) | A voice from a few seconds of a recording | 98 MB |
| Voice cloning (Chatterbox) | The studio-grade read, in nine files | About 1.1 GB |
| Cutout | Person, object, and the brush | 15 MB, 179 MB and 40 MB |
| Enhance | Picture restoration | 5 MB |
See Captions, voices and cutout for what each one does.
Platforms
Windows (x64 and ARM64), macOS (Apple Silicon and Intel), Linux (x86_64 and ARM64) and Android are supported. iOS and iPadOS builds are in beta. Same app, same project files, in 14 languages.