MoneyPrinterTurbo
harry0703/MoneyPrinterTurboReviewed by AI
What it does
MoneyPrinterTurbo turns a topic or keyword into a finished short video for TikTok, YouTube Shorts or Reels. An AI model writes the script, stock footage comes from Pexels or Pixabay (or your own clips), and it adds a voice-over, subtitles and background music. It can also upload the result. On Windows you unzip an 876 MB portable package and run start.bat; on Mac and Linux you install it with Python. You need a key for an AI model service (or a local model through Ollama) and a free stock-footage key. The basic Edge TTS voice needs no key.
How easy it is to install
B · installer plus one simple step · Windows
- Windows B
- macOS C
- Linux C
Windows is the easiest: an 876 MB portable package you unzip (in a folder path without spaces or special characters) and start with start.bat, then enter an AI model key and a stock-footage key in the web interface. Mac and Linux have no ready package, so you install Python 3.11, create a virtual environment and run a few commands (pip install mpt, mpt webui).
Numbers on GitHub
- Stars
- 136,917on Oct 31, 2026
- Downloads
- 158,6797,004 this week
- Growth
- +1,436stars in 5 days
Stars over time
Facts
- Runs on
- Windows, macOS, Linux
- You need to install
- Nothing extra
- To build from source
- Python, Docker, NVIDIA graphics card, FFmpeg
- Disk space
- 876 MB–2.6 GB · from the installer
- Accounts or keys
- AI model provider for scripts (OpenAI, Kimi, DeepSeek, Gemini, Claude and others; not needed with local Ollama or a script you write yourself), Pexels or Pixabay (free stock footage; not needed if you only use your own clips), Paid voice services (Azure Speech, ElevenLabs, MiniMax, Fish Audio, ModelBest VoxCPM and others), AI video generation services (MiniMax H3, Seedance, WaveSpeed, OFox, MuAPI), TikTok, Instagram or YouTube accounts for automatic upload
- Cost
- Free, API costs extra · The program is free. Script writing uses an AI model service, usually paid per use (a local model through Ollama is an option). Pexels and Pixabay keys are free, and the Edge TTS voice needs no key. AI-generated video clips from the listed services are paid.
- License
- MIT (Permissive)
- Can replace
- Pictory, AI, ElevenLabs
- Last update
- Oct 31, 2026
Risk signals (7)
- AttentionDependencies with known vulnerabilities
Pinned dependencies with known vulnerabilities (OSV): python-multipart: GHSA-v9pg-7xvm-68hf, PYSEC-2026-3040
automatic check
- Attentionbundled music unclear rights
README.md (背景音乐 section): "当前项目里面放了一些默认的音乐，来自于 YouTube 视频，如有侵权，请删除。" (the default background songs in resource/songs come from YouTube videos; delete them if they infringe). Videos published with these songs may get copyright claims.
AI review
- Infosponsor affiliate links
README.md has a long sponsor section of paid AI API platforms with referral links, for example "https://platform.kimi.com?track_id=track-6eec1e56a4494e52adcaebbcbbefce59&aff=moneyprinterturbo" and "https://infistar.cc/register?aff=6T4EYXP2". The model list in the features section also links to these sponsors.
AI review
- Infoupdates code at run time
README.md: "下载后，建议先双击执行 update.bat 更新到最新代码，然后双击 start.bat 启动" (run update.bat first to pull the latest code, then start.bat). The code you end up running can differ from the released archive.
AI review
- Infoagent installs from remote skill
README.md suggests giving an AI agent the remote file "https://raw.githubusercontent.com/harry0703/MoneyPrinterTurbo/main/docs/skill/SKILL.md"; the agent then installs, configures and runs the project in your terminal on its own.
AI review
- Infodocker build uses third party mirrors
Dockerfile defaults to ARG DOCKER_BUILD_MIRROR=china and, when building locally, writes Aliyun and Tsinghua Debian sources and runs "uv pip install --system --no-cache --default-index https://mirrors.aliyun.com/pypi/simple/". The prebuilt image used by docker-compose.release.yml is built with the official PyPI.
AI review
- Infosimilar name package on pypi
README.md: "官方 PyPI 包名为 mpt，由 harry0703 维护。PyPI 上名为 MoneyPrinterTurbo 的包由第三方维护。" (the official package is `mpt`; the PyPI package called `MoneyPrinterTurbo` is maintained by someone else).
AI review
Evidence is shown in its original language.
What Radar could not check
- Whether the portable Windows package starts cleanly and what update.bat changes after download.
- Real cost per video with each AI model, voice and video service.
- Disk use after the optional Whisper subtitle model downloads (README says about 3 GB for large-v3, 1.6 GB for large-v3-turbo).
- Whether automatic upload to TikTok, Instagram and YouTube Shorts works and follows those platforms' rules.
- The license of the bundled background music and of the stock clips that end up in the final videos.
- How good and how accurate the AI-written scripts and the matched footage are.