👻 GhostWrite: A Local AI Ghost Story Generator

Hacktoberfest Open-Source AI Challenge: Ghost in the Machine Submission 👻🌼

What happens when you combine a tiny open-weight language model with a little imagination?

For Hacktoberfest 2026's Open-Source AI Challenge, I built GhostWrite, a local AI-powered ghost story generator that turns simple prompts into atmospheric horror stories.

I wanted to build something fun, creative, and practical while exploring how open-source AI models can power applications without depending on proprietary AI APIs.

GhostWrite lets users provide three simple inputs:

The application combines these inputs to generate an original ghost story.

GhostWrite generating a story using custom user inputs.

A completed AI-generated ghost story, ready to download.

GhostWrite uses TinyLlama 1.1B Chat, an open-weight language model, through Hugging Face Transformers.

Using an open-weight model makes this project accessible to developers who want to experiment with AI without relying on paid API services.

Running inference locally also means that prompts can remain on the user's machine instead of being sent to an external AI API.

Open models make it easier to experiment with different prompts, modify generation parameters, and eventually replace the model with another suitable open-weight alternative.

This project is a small exploration of how open AI can make creative tools more accessible.

git clone https://github.com/AkshatMaheshwari27/ghostwrite cd ghostwrite Enter fullscreen mode Exit fullscreen mode Create and activate a virtual environment:

python -m venv.venv Enter fullscreen mode Exit fullscreen mode Install dependencies:

pip install -r requirements.txt Enter fullscreen mode Exit fullscreen mode Run the application:

streamlit run app.py Enter fullscreen mode Exit fullscreen mode The model downloads on the first run.

Built for the Hacktoberfest 2026 Open-Source AI Challenge: Week 4 — Ghost in the Machine.