Repo Image & Video

harry0703/MoneyPrinterTurbo

Open-source tool that turns a topic or keyword into an HD short video, with script, footage, voiceover, subtitles and music, via WebUI, API, CLI or AI Agent.

  • 129k GitHub stars
  • Python
  • ⚖️ MIT
  • 🎯 Beginner
View on GitHub (opens in a new tab) Setup guide
git clone https://github.com/harry0703/MoneyPrinterTurbo.git
harry0703/MoneyPrinterTurbo preview image

What it is

MoneyPrinterTurbo is a one-stop AI short-video generator. Given a video topic or keyword, it automatically produces a script, matches footage, generates subtitles and background music, and composes an HD video. It can be used through an AI Agent, a WebUI, an API or a CLI, and supports batch generation of multiple videos.

Who it's for

  • Creators who want to automatically generate short videos from a topic or keyword
  • Developers who want to plug video generation into automated workflows via the API or CLI
  • Users who prefer a no-install route (AI Agent, Google Colab, or the Windows one-click package)

Requirements

Requirements

  • Python 3.11 or higher for local deployment (3.11 recommended)
  • Supported systems: Windows 10, macOS 11.0 or later, and mainstream Linux distributions
  • Minimum 4-core CPU and 4 GB RAM; recommended 6-8 cores and 8 GB RAM; GPU not required
  • API keys for cloud LLMs, online footage or AI video services, entered in the WebUI basic settings
  • On Windows, avoid project paths with Chinese characters, special characters or spaces

Setup

  1. Windows one-click package

    Download the .7z archive from the latest release Assets (not the auto-generated Source code archives). Double-click update.bat to get the latest code, then double-click start.bat. The browser opens automatically; use Chrome or Edge if the page is blank.

  2. Clone the repository

    Clone the code. On first launch the project automatically creates config.toml from config.example.toml.

    bash
    git clone https://github.com/harry0703/MoneyPrinterTurbo.git

  3. Manual deployment with uv

    Install Python 3.11 and sync dependencies with the lock file.

    bash
    cd MoneyPrinterTurbo
uv python install 3.11
uv sync --frozen

  4. Start the WebUI (macOS/Linux)

    Run from the project root directory. On Windows run webui.bat instead.

    bash
    sh webui.sh

  5. Start the API service

    Starts the API server. API docs are available at http://127.0.0.1:8080/docs or /redoc.

    bash
    uv run python main.py

  6. Docker deployment

    Copy config.example.toml to config.toml before the first start. The release compose file pulls the prebuilt image from GitHub Container Registry. The WebUI is then at http://127.0.0.1:8501.

    bash
    cd MoneyPrinterTurbo
docker compose -f docker-compose.release.yml up

Examples

Generate a video with an AI Agent

Prompt
prompt
使用这个 Skill：https://raw.githubusercontent.com/harry0703/MoneyPrinterTurbo/main/docs/skill/SKILL.md
帮我生成一个主题为“人工智能如何改变普通人的日常生活”的视频。

Expected output: For agents that can read Skill docs and operate a local terminal. The agent installs, configures and generates the video, only asking for missing API keys, and returns the video file path. Supports macOS and Windows.

Generate a video from the command line

bash
bash
uv run python cli.py --video-subject "人工智能如何改变日常生活"

What it does: The simplest full video generation command, with no browser needed.

View all CLI options

bash
bash
uv run python cli.py --help

What it does: Shows the full command list, parameter descriptions and usage.

Run a batch of tasks

bash
bash
uv run python cli.py --batch-file ./tasks.json --stop-at video

What it does: Runs multiple tasks in sequence from a UTF-8 JSON array or JSONL manifest. CLI arguments act as global defaults and each object can override VideoParams fields. The manifest is limited to 100 tasks and 1 MiB, and one failed task does not block the rest.

Pros & cons

Pros

  • Pro:Covers the whole pipeline: script, footage, voiceover, subtitles, background music and editing, with custom content allowed at each step
  • Pro:Four usage modes (AI Agent, WebUI, API, CLI) plus Docker, Colab and a Windows one-click package
  • Pro:Broad provider support for LLMs, TTS (including free Edge TTS) and footage sources (local, Pexels, Pixabay, Coverr, AI text-to-video)
  • Pro:Supports 9:16, 16:9 and 1:1 output, batch generation, and automatic upload to TikTok, Instagram and YouTube Shorts

Cons

  • Con:Cloud LLM, footage and AI video features require you to supply API keys for the respective services
  • Con:Local transcription with faster-whisper, batch generation or heavy local processing benefit noticeably from a GPU
  • Con:AI Agent mode currently supports only macOS and Windows

Images

harry0703/MoneyPrinterTurbo preview image
harry0703/MoneyPrinterTurbo preview image · harry0703/MoneyPrinterTurbo on GitHub · MIT
harry0703/MoneyPrinterTurbo preview image