[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-74958":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":31,"readmeContent":32,"aiSummary":33,"trendingCount":16,"starSnapshotCount":16,"syncStatus":34,"lastSyncTime":35,"discoverSource":36},74958,"LTX-Desktop","Lightricks\u002FLTX-Desktop","Lightricks","An open-source desktop app for generating videos with LTX models","https:\u002F\u002Fwww.ltx.video",null,"TypeScript",1671,344,18,37,0,14,38,124,42,20.61,"Apache License 2.0",false,"main",true,[27,28,29,30],"generative-ai","ltx","ltx-2","non-linear-editing","2026-06-12 02:03:30","# LTX Desktop\n\nLTX Desktop is an open-source desktop app for generating videos with LTX models — locally on supported Windows\u002FLinux NVIDIA GPUs, with an API mode for unsupported hardware and macOS.\n\n> **Status: Beta.** Expect breaking changes.\n> Frontend architecture is under active refactor; large UI PRs may be declined for now (see [`CONTRIBUTING.md`](docs\u002FCONTRIBUTING.md)).\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"images\u002Fgen-space.png\" alt=\"Gen Space\" width=\"70%\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"images\u002Fvideo-editor.png\" alt=\"Video Editor\" width=\"70%\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"images\u002Ftimeline-gap-fill.png\" alt=\"Timeline gap fill\" width=\"70%\">\n\u003C\u002Fp>\n\n## Features\n\n- Text-to-video generation\n- Image-to-video generation\n- Audio-to-video generation\n- Video edit generation (Retake)\n- Video Editor Interface\n- Video Editing Projects\n\n## Local vs API mode\n\n| Platform \u002F hardware | Generation mode | Notes |\n| --- | --- | --- |\n| Windows + CUDA GPU with **≥16GB VRAM** | Local generation | Downloads model weights locally |\n| Windows (no CUDA, \u003C16GB VRAM, or unknown VRAM) | API-only | **LTX API key required** |\n| Linux + CUDA GPU with **≥16GB VRAM** | Local generation | Downloads model weights locally |\n| Linux (no CUDA, \u003C16GB VRAM, or unknown VRAM) | API-only | **LTX API key required** |\n| macOS (Apple Silicon builds) | API-only | **LTX API key required** |\n\nIn API-only mode, available resolutions\u002Fdurations may be limited to what the API supports.\n\n## System requirements\n\n### Windows (local generation)\n\n- Windows 10\u002F11 (x64)\n- NVIDIA GPU with CUDA support and **≥16GB VRAM** (more is better)\n- 16GB+ RAM (32GB recommended)\n- **160GB+ free disk space** (for model weights, Python environment, and outputs)\n\n### Linux (local generation)\n\n- Ubuntu 22.04+ or similar distro (x64 or arm64)\n- NVIDIA GPU with CUDA support and **≥16GB VRAM** (more is better)\n- NVIDIA driver installed (PyTorch bundles the CUDA runtime)\n- 16GB+ RAM (32GB recommended)\n- Plenty of free disk space for model weights and outputs\n\n### macOS (API-only)\n\n- Apple Silicon (arm64)\n- macOS 13+ (Ventura)\n- Stable internet connection\n\n## Install\n\n1. Download the latest installer from GitHub Releases: [Releases](..\u002F..\u002Freleases)\n2. Install and launch **LTX Desktop**\n3. Complete first-run setup\n\n## First run & data locations\n\nLTX Desktop stores app data (settings, models, logs) in:\n\n- **Windows:** `%LOCALAPPDATA%\\LTXDesktop\\`\n- **macOS:** `~\u002FLibrary\u002FApplication Support\u002FLTXDesktop\u002F`\n- **Linux:** `$XDG_DATA_HOME\u002FLTXDesktop\u002F` (default: `~\u002F.local\u002Fshare\u002FLTXDesktop\u002F`)\n\nModel weights are downloaded into the `models\u002F` subfolder (this can be large and may take time).\n\nOn first launch you may be prompted to review\u002Faccept model license terms (license text is fetched from Hugging Face; requires internet).\n\nText encoding: to generate videos you must configure text encoding:\n\n- **LTX API key** (cloud text encoding) — **text encoding via the API is completely FREE** and highly recommended to speed up inference and save memory. Generate a free API key at the [LTX Console](https:\u002F\u002Fconsole.ltx.video\u002F). [Read more](https:\u002F\u002Fltx.io\u002Fmodel\u002Fmodel-blog\u002Fltx-2-better-control-for-real-workflows).\n- **Local Text Encoder** (extra download; enables fully-local operation on supported Windows hardware) — if you don't wish to generate an API key, you can encode text locally via the settings menu.\n\n## API keys, cost, and privacy\n\n### LTX API key\n\nThe LTX API is used for:\n\n- **Cloud text encoding and prompt enhancement** — **FREE**; text encoding is highly recommended to speed up inference and save memory\n- API-based video generations (required on macOS and on unsupported Windows hardware) — paid\n- Retake — paid\n\nAn LTX API key is required in API-only mode, but optional on Windows\u002FLinux local mode if you enable the Local Text Encoder.\n\nGenerate a FREE API key at the [LTX Console](https:\u002F\u002Fconsole.ltx.video\u002F). Text encoding is free; video generation API usage is paid. [Read more](https:\u002F\u002Fltx.io\u002Fmodel\u002Fmodel-blog\u002Fltx-2-better-control-for-real-workflows).\n\nWhen you use API-backed features, prompts and media inputs are sent to the API service. Your API key is stored locally in your app data folder — treat it like a secret.\n\n### fal API key (optional)\n\nUsed for Z Image Turbo text-to-image generation in API mode. When enabled, image generation requests are sent to fal.ai.\n\nCreate an API key in the [fal dashboard](https:\u002F\u002Ffal.ai\u002Fdashboard\u002Fkeys).\n\n### Gemini API key (optional)\n\nUsed for AI prompt suggestions. When enabled, prompt context and frames may be sent to Google Gemini.\n\n## Architecture\n\nLTX Desktop is split into three main layers:\n\n- **Renderer (`frontend\u002F`)**: TypeScript + React UI.\n  - Calls the local backend over HTTP at `http:\u002F\u002Flocalhost:8000`.\n  - Talks to Electron via the preload bridge (`window.electronAPI`).\n- **Electron (`electron\u002F`)**: TypeScript main process + preload.\n  - Owns app lifecycle and OS integration (file dialogs, native export via ffmpeg, starting\u002Fmanaging the Python backend).\n  - Security: renderer is sandboxed (`contextIsolation: true`, `nodeIntegration: false`).\n- **Backend (`backend\u002F`)**: Python + FastAPI local server.\n  - Orchestrates generation, model downloads, and GPU execution.\n  - Calls external APIs only when API-backed features are used.\n\n```mermaid\ngraph TD\n  UI[\"Renderer (React + TS)\"] -->|HTTP: localhost:8000| BE[\"Backend (FastAPI + Python)\"]\n  UI -->|IPC via preload: window.electronAPI| EL[\"Electron main (TS)\"]\n  EL --> OS[\"OS integration (files, dialogs, ffmpeg, process mgmt)\"]\n  BE --> GPU[\"Local models + GPU (when supported)\"]\n  BE --> EXT[\"External APIs (only for API-backed features)\"]\n  EL --> DATA[\"App data folder (settings\u002Fmodels\u002Flogs)\"]\n  BE --> DATA\n```\n\n## Development (quickstart)\n\nPrereqs:\n\n- Node.js\n- `uv` (Python package manager)\n- Python 3.12+\n- Git\n\nSetup:\n\n```bash\npnpm setup:dev\n```\n\nRun:\n\n```bash\npnpm dev\n```\n\nDebug:\n\n```bash\npnpm dev:debug\n```\n\n`dev:debug` starts Electron with inspector enabled and starts the Python backend with `debugpy`.\n\nTypecheck:\n\n```bash\npnpm typecheck\n```\n\nBackend tests:\n\n```bash\npnpm backend:test\n```\n\nBuilding installers:\n- See [`INSTALLER.md`](docs\u002FINSTALLER.md)\n\n## Telemetry\n\nLTX Desktop collects minimal, anonymous usage analytics (app version, platform, and a random installation ID) to help prioritize development. No personal information or generated content is collected. Analytics is enabled by default and can be disabled in **Settings > General > Anonymous Analytics**. See [`TELEMETRY.md`](docs\u002FTELEMETRY.md) for details.\n\n## Docs\n\n- [`INSTALLER.md`](docs\u002FINSTALLER.md) — building installers\n- [`TELEMETRY.md`](docs\u002FTELEMETRY.md) — telemetry and privacy\n- [`backend\u002Farchitecture.md`](backend\u002Farchitecture.md) — backend architecture\n\n## Contributing\n\nSee [`CONTRIBUTING.md`](docs\u002FCONTRIBUTING.md).\n\n## License\n\nApache-2.0 — see [`LICENSE.txt`](LICENSE.txt).\n\nThird-party notices (including model licenses\u002Fterms): [`NOTICES.md`](NOTICES.md).\n\nModel weights are downloaded separately and may be governed by additional licenses\u002Fterms.\n","LTX Desktop 是一款开源桌面应用程序，用于利用LTX模型生成视频。它支持文本到视频、图像到视频、音频到视频的转换以及视频编辑功能，并提供了一个直观的视频编辑界面。该应用基于TypeScript开发，能够在支持CUDA的Windows和Linux系统上本地运行，对于不支持的硬件或macOS用户，则需要通过API模式使用，此时需提供LTX API密钥。适用于需要快速创作多媒体内容的个人创作者或小型团队，特别是在拥有强大GPU资源的情况下能够发挥最佳性能。此外，项目正处于Beta阶段，持续进行前端架构重构以优化用户体验。",2,"2026-06-11 03:51:44","high_star"]