Generate a short video from a text prompt or a single image.
Run AI video generation locally on a powerful NVIDIA GPU without sending data to the cloud.
Use the cloud API to generate videos on a Mac or a weaker Windows or Linux machine.
Edit and arrange generated video clips on a timeline inside the built-in editor.
| lightricks/ltx-desktop | trekhleb/js-image-carver | lingyichen-ai/jadeai | |
|---|---|---|---|
| Stars | 1,583 | 1,596 | 1,619 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | hard | easy | moderate |
| Complexity | 4/5 | 2/5 | 2/5 |
| Audience | vibe coder | developer | general |
Figures from each repo's GitHub metadata at analysis time.
Local generation needs an NVIDIA GPU with 16GB+ VRAM and over 160GB of free disk space for model weights, otherwise an LTX API key is required.
LTX Desktop is a free desktop app from Lightricks for generating videos using the company's LTX AI models. It supports turning a text description into a video, turning a single image into a video, turning audio into a video, and editing an existing video generation, along with a built in video editor interface for arranging clips on a timeline. How the app generates video depends on your computer. On Windows or Linux with a supported NVIDIA graphics card that has at least 16 gigabytes of video memory, it can download the model and run generation entirely on your own machine. On Mac computers, or on Windows and Linux machines without a strong enough NVIDIA card, it instead relies on Lightricks' cloud API to generate videos, which requires a free account and an API key, though the video generation itself through the API is a paid service. Running fully locally on a supported machine also needs a large amount of free disk space, since the model weights themselves are large, plus a reasonable amount of system memory. Even in local mode, you still need to set up how text prompts get converted for the model, either using the free cloud based text encoding through an API key, which is recommended because it is faster and uses less memory, or downloading an additional local text encoder to avoid using the API at all. There are also two optional integrations: one with fal.ai for a specific image generation feature, and one with Google's Gemini for AI suggested prompt improvements, both of which only send data externally if you choose to enable them. Under the hood, the app is split into a React and TypeScript interface, an Electron layer that handles the desktop app shell and file operations, and a Python backend using FastAPI that manages model downloads and runs the actual generation. The app collects minimal anonymous usage statistics by default, which can be turned off in settings, and it is released under the Apache 2.0 license, though the downloaded AI model weights themselves may carry separate license terms.
LTX Desktop is a free desktop app for generating videos from text, images, or audio using Lightricks's LTX AI models, running locally on strong NVIDIA GPUs or via a cloud API otherwise.
Mainly TypeScript. The stack also includes TypeScript, React, Electron.
You can use, modify, and distribute this software freely, including for commercial purposes, as long as you follow the Apache 2.0 license's attribution and notice requirements, downloaded AI model weights may have separate license terms.
Setup difficulty is rated hard, with roughly 1h+ to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.