MoneyPrinter V2

1. What is MoneyPrinter V2?

MoneyPrinter V2 (MPV2 for short) is a command-line program that automates a handful of "make money online" chores. It has four functions: creating and uploading YouTube Shorts, posting to Twitter (X) on a schedule, affiliate marketing for Amazon products through Twitter, and finding local businesses to send outreach emails to.

For YouTube Shorts, the program uses AI to pick a topic, write a script, generate illustrations, read the script aloud, assemble a vertical video and upload it to your channel. Script writing uses an AI model that runs on your own computer through Ollama, so it costs nothing. Image generation uses a Google service and needs an API key.

Three things you should know up front:

• This is a tool for people comfortable with the command line. There is no installer and no window with buttons. You work by typing numbers into a text menu.
• Despite the name, it does not guarantee any income. The author states that the project is for educational purposes only and that you use it at your own risk.
• It is free and open source under the AGPL-3.0 license.

A good fit if:

• You want to experiment with automated short videos and enjoy tinkering.
• Your computer is strong enough to run an AI model locally.
• You can read English status messages.

Not a good fit if:

• You have never used a terminal and want something that works with one click.
• You only have a phone.
• You need voice-over in a language other than English: the built-in voices are the KittenTTS ones (Bella, Jasper, Luna, Bruno, Rosie, Hugo, Kiki, Leo), and the documentation does not mention other languages.

2. What do you need first?

Computer: Windows, macOS or Linux. The project publishes no minimum requirements. In practice the machine has to run a language model through Ollama and render video, so a weak computer will be very slow.

Software to install beforehand:

Software	What it is	Where to get it
Python 3.12	The language the program is written in. The README asks for version 3.12 specifically	https://www.python.org/downloads/
Git	Downloads source code	https://git-scm.com/downloads
Ollama	Runs text-generating AI models on your own machine, free	https://ollama.com/download
Firefox	The program drives Firefox to post on your behalf	https://www.mozilla.org/firefox/
ImageMagick	An image tool, used to draw subtitles onto the video	https://imagemagick.org/script/download.php
Go	Only needed for the email outreach function	https://go.dev/dl/

Accounts and keys:

• A Google Gemini API key for image generation (the config calls it "Nano Banana 2"). Without it the preflight check reports a blocking failure. Whether it costs money depends on Google's terms when you use it; check Google's official page.
• A YouTube or Twitter (X) account already signed in inside Firefox.
• Optional: an AssemblyAI key if you want subtitles from an outside service rather than Whisper running locally; a Post Bridge account if you want cross-posting to TikTok and Instagram.

3. Step-by-step installation

The project has no packaged installer. The command line is the only route.

Step 1: Install the prerequisites

Install Python 3.12, Git, Ollama, Firefox and ImageMagick from the links in section 2. When installing Python on Windows, tick "Add python.exe to PATH".

Check them:

python --version
git --version
ollama --version

What success looks like: each command prints a version number, and Python reports 3.12.

Step 2: Download a model for Ollama

ollama pull llama3.2:3b

This is the model the program itself suggests in its preflight message. What success looks like: the progress bar reaches 100% and the command ends without an error.

Step 3: Download the source and create the config file

git clone https://github.com/FujiwaraChoki/MoneyPrinterV2.git
cd MoneyPrinterV2
cp config.example.json config.json

On Windows, cp works in PowerShell. What success looks like: the MoneyPrinterV2 folder now contains a config.json file.

Step 4: Create a virtual environment and install the packages

A virtual environment is a private folder for this project's Python packages, so they do not mix with anything else on your machine.

python -m venv venv

Activate it on Windows:

.\venv\Scripts\activate

Activate it on macOS and Linux:

source venv/bin/activate

What success looks like: your prompt now starts with (venv). Then install the packages:

pip install -r requirements.txt

This downloads quite a lot and can take several minutes.

Note: On macOS and Linux the repository includes a script that combines Steps 3 and 4, detects the ImageMagick path, the Firefox profile (on macOS) and an Ollama model, then runs the preflight check: bash scripts/setup_local.sh.

Step 5: Fill in config.json

Open config.json in a text editor and set these entries:

Entry	What to put
firefox_profile	The path to the Firefox profile folder that is signed in to YouTube or Twitter. Type about:profiles into Firefox's address bar to find it
imagemagick_path	On Windows: the path to magick.exe. On macOS and Linux: the path to convert, usually /usr/bin/convert
nanobanana2_api_key	Your Gemini API key
ollama_model	The model you pulled, for example llama3.2:3b. Leave it empty and the program asks you to pick one at startup
headless	false to watch the Firefox window while it works; true to keep it hidden

The rest can stay as it is. Every entry is explained in docs/Configuration.md in the repository.

Note: In a JSON file on Windows, backslashes in paths must be doubled, for example C:\\Program Files\\.... A single backslash breaks the file.

Step 6: Run the preflight check

python scripts/preflight_local.py

What success looks like: lines starting with [OK] and a final "Preflight passed. Local setup looks ready." A [FAIL] line is a problem you must fix; a [WARN] line is a warning. See section 5.

4. First use

Make sure Ollama is running and that you are in the root MoneyPrinterV2 folder with the virtual environment active. Run:

python src/main.py

The program prints a menu:

1. YouTube Shorts Automation
2. Twitter Bot
3. Affiliate Marketing
4. Outreach
5. Quit

To make a first Short:

1. Type 1 and press Enter.
2. With no account stored yet, the program asks whether to create one. Answer Yes, then enter a nickname for the account, the Firefox profile path, the channel's niche and its language.
3. Select the account you just created. A new menu appears: "Upload Short", "Upload Short with custom topic", "Show all Shorts", "Setup CRON Job", "Quit".
4. Choose "Upload Short with custom topic" and type your video idea. With plain "Upload Short", the AI comes up with a topic based on the channel's niche.
5. Wait. The program writes the script, generates the images, records the voice and assembles the video. When it is done it asks "Do you want to upload this video to YouTube? (Yes/No)".

What the result looks like: a vertical video made of AI-generated images with a voice-over and background music. Answer Yes and Firefox opens and uploads it to the signed-in channel.

"Setup CRON Job" lets you schedule automatic uploads once, twice or three times a day.

Note: Watch the video before you publish it. Mass-posting automated content can break a platform's rules and get an account restricted. Try it on a secondary account first.

5. Common problems and fixes

Symptom	Cause	Fix
[FAIL] Ollama is not reachable at http://127.0.0.1:11434	Ollama is not running	Start the Ollama app and try again. Check ollama_base_url in config.json
[WARN] No models found on Ollama. Pull a model first	No model has been downloaded	Repeat the command in Step 2
[FAIL] nanobanana2_api_key is empty (and GEMINI_API_KEY is not set)	The Gemini key is missing	Put the key in nanobanana2_api_key, or set the GEMINI_API_KEY environment variable
[WARN] imagemagick_path is not set to a valid executable path	The ImageMagick path is wrong, so subtitles may fail to render	Point imagemagick_path at magick.exe (Windows) or convert (macOS, Linux)
[WARN] firefox_profile is empty or firefox_profile does not exist	The Firefox profile folder is missing or mistyped	Copy the correct path from Firefox's about:profiles page
[FAIL] Missing config file	config.json was never created	Repeat the cp command in Step 3
pip install fails while installing packages	Usually Python is not version 3.12	Install Python 3.12, delete the venv folder and redo Step 4
Windows refuses to run the script when activating venv	PowerShell blocks script files	Open Command Prompt (cmd) instead of PowerShell and run venv\Scripts\activate.bat
"python: command not found" or "git is not recognized"	Not installed, or not added to PATH	Reinstall with the PATH option ticked and reopen the terminal. On macOS and Linux try python3
python src/main.py says the file cannot be found	You are in the wrong folder	Use cd to return to the root MoneyPrinterV2 folder and run it again

If your problem is not listed, search the repository's Issues: https://github.com/FujiwaraChoki/MoneyPrinterV2/issues.

6. Uninstalling and updating

Updating: the project does not publish numbered releases (its Releases page is empty); new code goes straight to the main branch. The most recent change at the time of writing was on 15 September 2026. To get the latest code, go into the MoneyPrinterV2 folder with the virtual environment active:

git pull
pip install -r requirements.txt

After updating, compare your config.json with config.example.json in case new entries were added.

Removing:

1. If you scheduled automatic posting, stop the running program first.
2. Delete the whole MoneyPrinterV2 folder. It contains the venv environment, the config.json file with your keys, and the hidden .mp folder where the program keeps its working data.
3. If you no longer need it, remove the Ollama model with Ollama's standard command:

ollama rm llama3.2:3b

4. Python, Git, Ollama, Firefox and ImageMagick are removed like any other application.
5. Revoke the Gemini API key in your Google account if you do not need it.

7. Frequently asked questions

Does it cost anything? The program is free. Script writing (Ollama), the voice (KittenTTS) and subtitles (local Whisper) all run on your machine at no charge. Image generation calls Google's Gemini API with your key and is billed according to Google's pricing. AssemblyAI and Post Bridge are optional outside services.

Do I need an internet connection? Yes. It is needed to generate images through Google, download background music and post to YouTube or Twitter.

Is my data sent anywhere? Scripts are generated on your own machine. Image prompts are sent to Google to produce the pictures. The program uses your Firefox profile, meaning your actual signed-in YouTube or Twitter session, so do not point it at a profile holding accounts you cannot afford to lose.

Will I actually make money? Nothing guarantees it. The author states that the project is for educational purposes only and accepts no responsibility for how it is used. Treat it as an experiment.

Will it run on a weak computer? The project lists no specific requirements. You can choose a small Ollama model and leave whisper_model at its default base to keep things light. The weaker the machine, the longer each video takes.