67K+ Stars on GitHub: Automated Faceless Video Generation via MoneyPrinterTurbo for Beginners

You want a side hustle on TikTok, YouTube Shorts, or Instagram Reels—but you do not want to learn Premiere, show your face, or spend three hours per clip. MoneyPrinterTurbo (MIT license, 67,000+ GitHub stars as of May 2026) turns a single topic or keyword into a finished vertical or horizontal short: AI script, stock footage, voice-over, subtitles, background music, and final render—mostly unattended.

This is a faceless video factory for pure beginners: you pick the niche, the tool handles the mechanical pipeline. Streamlit Web UI on port 8501, optional REST API on 8080. China-based creators often pair DeepSeek or Moonshot LLMs. On macOS, batch mode shines on an always-awake Mac—desk Mac mini or hosted macOS—for overnight renders. For multi-agent content ops (research → CSV → draft), see our Hermes subagent delegation guide; this article focuses on video output, not text pipelines.

How MoneyPrinterTurbo works internally

MoneyPrinterTurbo uses a clear MVC layout (per upstream README). Data flows in one direction:

Topic / keyword
    → LLM script (config.toml llm_provider + API keys)
    → Stock footage (Pexels API) or local clips
    → TTS voice (edge / Azure)
    → Subtitles (edge or whisper ~3GB)
    → Background music (resource/songs/)
    → ffmpeg + ImageMagick → MP4 (9:16 or 16:9)

Quotable: Upstream recommends 8 GB RAM minimum, 16 GB for smoother batch runs, and 4 GB+ GPU VRAM only for heavier local whisper—not required with cloud LLMs and edge subtitles.

Source: MoneyPrinterTurbo README.

Platform matrix: where faceless Shorts fit

If you target Shorts only, do enable batch generation and pick the best of 3–5 renders. If you need talking-head credibility, do not expect this tool to replace on-camera builds—it excels at stock B-roll + voice-over faceless formats.

Step-by-step macOS runbook

Eight steps from clone to optional API automation—tested on Apple Silicon and Intel Macs you control.

1. Clone the repo — Avoid Chinese characters or spaces in the install path (upstream warning).

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

2. Install dependencies with uv — Python 3.11 is the upstream default.

uv python install 3.11
uv sync --frozen

3. Install ImageMagick — Required for subtitle rendering on macOS (ffmpeg via Homebrew).

brew install imagemagick ffmpeg

4. Configure API keys — Copy config.example.toml → config.toml. Set pexels_api_keys (free tier at pexels.com/api).
5. Launch the Web UI — Open http://127.0.0.1:8501 in Chrome or Edge if the auto tab is blank.

uv run streamlit run ./webui/Main.py --browser.gatherUsageStats=False

6. Generate your first vertical short — Enter a topic (e.g. "3 habits that improve sleep"), choose 9:16, pick voice + subtitle style, click generate.
7. Batch mode for matrix publishing — Generate 3–5 MP4s with varied hooks; schedule on a Mac that does not sleep (pmset -c sleep 0 on plugged-in laptops).
8. Optional API automation — Start uv run python main.py and hit http://127.0.0.1:8080/docs—wrap inside a Hermes delegate_task worker that drops topics from a spreadsheet.

Hardware reference: Apple Mac mini specifications list 16 GB on base M4—enough for Web UI + one render; choose 24 GB with whisper locally plus browser tabs.

Troubleshooting

RuntimeError: No ffmpeg exe could be found

Symptom: Render fails immediately; stack trace mentions IMAGEIO_FFMPEG_EXE.

Fix: Install ffmpeg (brew install ffmpeg) or download a static build from gyan.dev ffmpeg builds:

[app]
ffmpeg_path = "/opt/homebrew/bin/ffmpeg

ImageMagick policy blocks temporary text files

Symptom: Error referencing pattern="@" and policy.xml.

Fix: Edit ImageMagick policy.xml (Homebrew: /opt/homebrew/etc/ImageMagick-7/policy.xml)—change the @ pattern from rights="none" to rights="read|write".

OSError: Too many open files during batch render

Symptom: Batch job dies mid-queue on macOS.

Fix: Raise the limit before launching Streamlit:

ulimit -n 10240
uv run streamlit run ./webui/Main.py --browser.gatherUsageStats=False

Whisper model download fails (China network)

Symptom: HuggingFace timeout when subtitle_provider = whisper.

Fix: Stay on edge subtitles first. If you need whisper, manually download whisper-large-v3 (~3 GB) per upstream mirrors into MoneyPrinterTurbo/models/whisper-large-v3/.

Side-hustle workflow: topic → batch → publish

A repeatable solopreneur loop for faceless channels:

1. Monday — Brainstorm 10 keywords in a spreadsheet (Notion or CSV).
2. Tuesday — Batch-generate 5 vertical MP4s; pick top 3 by manual preview.
3. Wednesday–Friday — Upload one per day to TikTok / YouTube Shorts / Instagram Reels; reuse hooks as carousel or thread captions.
4. Weekend — Review analytics; feed winning topics back into next week's batch.

Pair with automation: Hermes subagent Worker A researches trending hooks, Worker B fills the CSV, you paste topics into MoneyPrinterTurbo—or call the REST API from a cron script on an always-on Mac.

Related reading

Related reading: Hermes subagent delegation guide for research → CSV → draft pipelines that complement faceless video output—no pricing tables here.

FAQ

Is MoneyPrinterTurbo really free?

The software is MIT-licensed and free to self-host. You still pay for LLM API usage, optional Azure TTS, and Pexels (free tier has rate limits). Budget roughly $5–20/month in API credits for casual batch publishing—not zero, but far below hiring an editor.

Can I use it without showing my face?

Yes—that is the core use case. Output is stock footage + synthesized voice + burned-in subtitles. Viewers never see you; focus on niche selection and hook quality instead of lighting and B-roll shooting.

TikTok vs YouTube Shorts—which settings?

Both prefer 9:16 at 1080x1920. TikTok tolerates slightly longer clips in some markets; YouTube Shorts must stay under 60 seconds for Shorts shelf placement—shorten segment duration in the Web UI or trim in upload.

Do I need a GPU on Mac mini M4?

No for the default cloud-LLM + edge-subtitle path. GPU helps only if you switch to local whisper subtitles or run many parallel ffmpeg jobs; upstream lists 4 GB VRAM as recommended, not required.

How does this compare to online AI video generators?

Upstream thanks RecCloud for hosted no-install access. Self-hosting MoneyPrinterTurbo gives unlimited batch control, local file ownership, and API hooks—trade-off is setup time. Hosted tools win if you refuse to touch a terminal; compare https://reccloud.com if you want zero install.