LTX Desktop system requirements
LTX Desktop runs on Windows, macOS, and Linux. Whether generation runs locally or through the LTX API depends on your hardware — check the requirements for your platform before installing.
Windows
For local generation:
- Windows 10 or 11 (64-bit)
- NVIDIA GPU (32GB+ VRAM recommended)
- 160GB+ free disk space (for model weights, Python environment, and outputs)
- Internet connection for initial model download
Local generation may work on machines with 16GB VRAM using in-app optimizations, but 32GB+ is the recommended spec for reliable performance.
However, Windows machines without a qualifying NVIDIA GPU run in API mode instead.
macOS
- Apple Silicon (M1 or later)
- macOS 13 (Ventura) or later
- 16GB+ unified memory
- Stable internet connection
Generation on macOS uses LTX API mode — local GPU inference is not currently supported on Mac. An LTX API key is required.
Linux
For local generation:
- Ubuntu 22.04+ or similar distribution (x64 or arm64)
- NVIDIA GPU (32GB+ VRAM recommended)
- NVIDIA drivers installed
- 160GB+ free disk space
- Internet connection for initial model download
Local generation may work on machines with 16GB VRAM using in-app optimizations, but 32GB+ is the recommended spec for reliable performance.
However, Linux machines without a qualifying NVIDIA GPU run in API mode.
AMD and Intel GPUs
AMD and Intel GPUs are not currently supported for local inference. Machines with these GPUs will use API mode.
After installing, LTX Desktop detects your hardware automatically. If only API models appear in the generation panel, your machine is running in API mode. See Local mode and API mode in LTX Desktop for a full explanation of how the two modes work and what to expect on your setup.