ShortGPT
VerifiedAI framework that automates short video production from script to render.
What is ShortGPT?
ShortGPT is an open-source framework that turns natural-language instructions into finished short videos. It manages the full pipeline: generating scripts, sourcing stock footage, synthesizing voiceovers, adding captions, and performing edits using a custom LLM-oriented editing language.
The system stores state with TinyDB for persistence across runs and exposes a Gradio web UI for local control. Users can run it via Docker or a hosted Colab notebook, making it accessible without heavy local setup.
It is aimed at YouTube and TikTok creators, marketers, and developers who need to produce multilingual short-form content on a regular schedule without manual editing.
What you can build with ShortGPT
YouTube Shorts automation
Generate, voice, and edit daily short videos from topic prompts with minimal oversight.
TikTok content pipeline
Create trending clips complete with captions and background music using automated asset sourcing.
Multilingual voice dubbing
Translate and redub existing videos into 40+ languages while preserving timing and captions.
Install ShortGPT
docker build -t short_gpt_docker:latest . && docker run -p 31415:31415 --env-file .env short_gpt_docker:latestdocker build -t short_gpt_docker:latest .
docker run -p 31415:31415 --env-file .env short_gpt_docker:latest- 1Clone the ShortGPT repository and prepare a .env file with required API keys.
- 2Build the Docker image with the provided Dockerfile.
- 3Run the container on port 31415 using the docker run command.
- 4Open http://localhost:31415 in a browser to access the Gradio interface.
- 5Enter a content prompt and monitor the automated video generation steps.
ShortGPT: pros & cons
Pros
- +End-to-end automation from idea to rendered video
- +Broad language support including many non-English options
- +Open-source with Docker and Colab deployment paths
- +Built-in memory for long-running projects
Cons
- –Requires Docker and external API keys for full functionality
- –Local setup can be involved for non-technical users
- –Asset quality depends on third-party stock libraries
Frequently asked questions
No GPU is required for basic operation, though faster inference benefits from one when using local LLMs.
User reviews
Verified reviews from the community shape this listing's rating.
Loading reviews…