
MoneyPrinterTurbo
Turn a single topic or keyword into a finished HD short video, end-to-end.
Solo creators, growth marketers and dev-savvy content studios who want to mass-produce faceless TikTok/Reels/Shorts on a self-hosted, provider-agnostic pipeline they can script and extend.
Filmmakers or brands who need bespoke cinematography, licensed talent or original AI-generated footage - MoneyPrinterTurbo composites stock clips, it does not generate video frames.
MoneyPrinterTurbo is an open-source Python pipeline that automates the entire short-form video assembly line. Give it a topic or keyword and it will draft a script with an LLM, generate matching keywords, pull royalty-free stock footage from Pexels, Pixabay or Coverr (or use your own local clips), synthesize voice-over via Edge TTS, Azure Speech, SiliconFlow, ElevenLabs, Google Gemini or Chatterbox, cut and align the clips, burn in stylable subtitles, mix in background music, and render vertical 1080x1920 or horizontal 1920x1080 MP4s ready for TikTok, Instagram Reels or YouTube Shorts. It ships four front doors to the same engine, a Streamlit WebUI, a REST API, a CLI, and an AI-Agent Skill you can hand to Claude or another coding agent to run the whole install and render for you. The LLM layer is provider-agnostic: it plugs into Kimi/Moonshot, OpenAI, DeepSeek, Google Gemini, Qwen, Azure OpenAI, ByteDance Volcengine Ark, xAI Grok, MiniMax, plus gateway/aggregator layers like Cloudflare AI Gateway, ModelScope, OneAPI, LiteLLM, Groq and local Ollama, so you can trade off cost, latency and censorship rules per market. Batch mode lets creators queue many variants in one shot and pick the best cut. The project is a top-1000 GitHub repo (over 100k stars) actively maintained by harry0703, with Docker, uv and Windows one-click launcher installs, plus a hosted zero-install version by third party Reccloud for non-technical users.
The most complete open-source short-video assembly line we have tested. It will not win Cannes, but if your job is to publish 30 faceless shorts a week across niches, this is the pipeline to fork - the multi-LLM abstraction and batch mode alone save the setup cost of a boutique tool, and the MIT license means your automation is not one Terms-of-Service update away from dying.
— The AI Tool Bible editorial team
Pros
- ✅ Fully open source (MIT) so you own the pipeline and can self-host without per-video fees
- ✅ Model-agnostic LLM layer supports 15+ providers plus local Ollama, avoiding vendor lock-in
- ✅ Four interfaces (WebUI, REST API, CLI, AI-Agent Skill) make it embeddable in almost any workflow
- ✅ Handles the full stack: script, keywords, footage retrieval, TTS, subtitles, BGM, render
- ✅ Batch generation lets you crank out many variants of the same topic and pick the winner
- ✅ Wide TTS coverage including Edge TTS (free), Azure, ElevenLabs and Chatterbox with live preview
- ✅ Runs on modest hardware; GPU only needed if you want local Whisper transcription or faster batches
Cons
- ⚠️ Output quality is capped by stock-footage relevance; videos can feel generic or loosely on-topic
- ⚠️ Requires Python 3.11+, API keys and some troubleshooting; non-devs will struggle without the hosted mirror
- ⚠️ No native AI video generation (Sora/Veo/Kling style) - it composites existing stock clips, not new footage
- ⚠️ Primary docs and issue threads are in Chinese; English README is a translation and lags behind
- ⚠️ Because it targets high-volume short-form output, misuse for low-effort spam content on TikTok/YouTube is a real risk
- ⚠️ Voice-over/stock-clip alignment can drift on longer scripts and often needs manual editing
Use cases
Explore related
Compare with similar tools
All in Video →
Runway
FeaturedPro-grade AI video editor and Gen-4 generation.

Sora
FeaturedOpenAI's flagship text-to-video model.

Luma Dream Machine
Fast, accessible text-to-video with strong camera control.

HeyGen
Avatar video + lip-sync translation at scale.

Google Veo
Google DeepMind's flagship text-to-video model with native audio generation and cinematic camera control.

Higgsfield
AI video and image generation suite that aggregates 30+ frontier models under one workflow.