Join waitlist
Back to products
LTX Desktop

LTX Desktop

Local open-source LTX video editor optimized for GPUs

Website ltx.io
Overview

What it is

LTX Desktop combines a full non-linear video editor with on-device AI generation. Free, open-source, runs locally on your machine. Powered by LTX-2.3.

Intent

I need it when

Maintain full control over creative IP and avoid external data transfers

LTX Desktop keeps all footage, prompts, and outputs on the user's machine with no forced cloud uploads. The open-source Apache 2.0 license allows forking and on-premises deployment, giving enterprises full architectural transparency and IP protection.

Generate and edit videos locally without cloud dependency or per-generation fees

LTX Desktop runs entirely on local hardware with zero usage fees, enabling text-to-video, image-to-video, audio-to-video, and retake capabilities. Users retain full data privacy and avoid recurring API costs by processing on their own machine.

Create professional video content with integrated AI generation and non-linear editing in one tool

LTX Desktop combines a full non-linear video editor with on-device AI generation, supporting professional editing tools (trim, transitions, color correction), timeline import from Premiere Pro/DaVinci Resolve/Final Cut Pro, and AI-native features like context-aware gap fill and non-destructive retakes.

Use open-source AI video tools for free without licensing restrictions for smaller organizations

LTX Desktop is fully open-source under Apache 2.0 and free for individuals and companies under $10M annual revenue. Users can download, modify, and deploy the model weights locally without per-generation costs or commercial licensing fees.

Iterate quickly on video concepts with multiple AI-generated takes without leaving the timeline

LTX Desktop enables generating multiple takes per clip directly in the timeline and switching between them non-destructively. Each version nests within the clip, keeping the timeline modular and supporting rapid creative exploration without re-rendering.

Drop

Not a fit when

  • User has incompatible GPU hardware (requires NVIDIA RTX with 32GB+ VRAM for Windows; macOS requires API mode which incurs costs)
  • User needs real-time cloud-based generation without local infrastructure setup and model downloads
  • User requires commercial licensing for companies with annual revenue above $10M without negotiating custom terms
  • User operates on macOS and wants local GPU inference (currently only API-based generation supported on macOS)
  • User needs immediate production without 160GB+ disk space for model downloads and setup time
Commercials

Pricing

Free for local inference on compatible hardware; optional API-based inference available via LTX API with usage-based pricing View pricing