Home

2 min read Original article ↗

LTX Video Generator

Transform text into stunning AI-generated videos on your Mac.

Download Latest Release View on GitHub


Native macOS Experience

LTX Video Generator is a beautiful, native macOS application built with SwiftUI. It runs the LTX-2 model natively on Apple Silicon using MLX (Apple’s machine learning framework) for optimal performance.

Key Features

  • Apple Silicon Native - Uses MLX for optimal M1/M2/M3/M4 performance
  • LTX-2 Model - 19B parameter video generation model from Lightricks
  • Text-to-Video - Generate videos from text descriptions
  • Image-to-Video - Animate images into videos
  • Voiceover Narration - Add TTS audio using ElevenLabs (cloud) or MLX-Audio (local)
  • Background Music - 54 genre presets for AI-generated instrumental music via ElevenLabs
  • Auto Package Installer - Missing Python packages detected and installed with one click
  • Generation Queue - Queue multiple videos and track progress in real-time
  • Smart History - Browse, preview, and manage all your generated videos
  • Flexible Presets - Quick access to common configurations or customize every parameter

Quick Start

  1. Download the app from the Releases page
  2. Open Preferences and click Auto Detect to find Python
  3. Install packages if prompted (one-click install available)
  4. Generate your first video! (model downloads on first run)

System Requirements

Requirement Minimum Recommended
macOS 14.0+ 15.0+
Processor Apple M1 Apple M2 Pro/M3/M4
Unified Memory 32GB 64GB+
Storage 100GB free 150GB+ free
Python 3.10+ 3.12+

First Run Download: The LTX-2 model (~90GB) downloads automatically on first generation. This is a one-time download cached in ~/.cache/huggingface/.

Sample Results

Generate videos like:

  • “A river flowing through a misty forest at dawn”
  • “The camera slowly pans across a futuristic cityscape”
  • “Golden leaves falling in slow motion against a blue sky”

Getting Help

Contributing

LTX Video Generator is open source! Contributions, issues, and feature requests are welcome on GitHub.