Automatically generate a finished YouTube Shorts video on any topic without filming or editing anything yourself.
Run a batch queue of video generation jobs using the Docker-backed API and worker so jobs survive restarts.
Produce short-form social media videos using a locally running AI model with no per-request API fees.
Requires Docker, ImageMagick configured on your system path, and Ollama running with a downloaded language model before the first video can be generated.
MoneyPrinter is a Python tool that automatically creates short videos for YouTube Shorts, the format of brief vertical clips. The idea is that you give it a topic and it builds a finished video without you filming or editing anything yourself. The playful name hints at the goal many users have, which is producing a steady stream of social media content with little manual effort. The tool relies on AI to do the writing. It is built around Ollama, a program that runs AI language models directly on your own computer rather than through a paid online service. After you start Ollama and download a model, MoneyPrinter uses it to generate the video's script and the accompanying metadata such as titles. Because the model runs locally, this part does not depend on an external AI provider. For assembling the actual video, the project uses MoviePy, a Python library for stitching together clips and text, and it needs a separate image tool called ImageMagick installed on your system. Recent versions changed how the work is processed. MoneyPrinter now runs through a database backed queue made up of an API, a worker, and a PostgreSQL database, all packaged with Docker. The README explains this makes generation more reliable and able to survive restarts, so a job is not lost if the program stops partway. There is also a user interface where you pick which AI model to use. The README is mostly a short FAQ covering common setup snags. It explains how to obtain a TikTok session ID by copying a browser cookie, what to do if ImageMagick is not detected automatically, and how to work around a failed installation of an audio library. Documentation for quickstart, configuration, architecture, Docker, and troubleshooting lives in a separate docs folder, and there is a Discord community for questions. A few practical notes round it out. The project is sponsored by a service called Post Bridge, it accepts donations and lists supporters, and the author states that pull requests are not being accepted at this time. Licensing details are kept in a separate license file.
← fujiwarachoki on gitmyhub — every repo by this author, as a profile.
Verify against the repo before relying on details.