Project Overview

The AI Story Generator is an innovative application that leverages local Large Language Models (LLMs) to create engaging and creative stories based on user prompts. Unlike many AI story generators that rely on external APIs like OpenAI, this application uses Ollama to run LLM models locally, providing greater privacy and control over the generation process.

Key Features

• User-friendly Streamlit interface for easy interaction
• Integration with Ollama to run local LLM models
• Customizable story parameters (length, style, genre)
• No dependency on external APIs, enhancing privacy
• Real-time story generation with minimal latency
• Support for various storytelling formats and styles

Technologies Used

• Python: Core programming language
• Streamlit: Frontend framework for the user interface
• Ollama: Tool for running local LLM models
• LLM Models: Various language models for story generation
• HTML/CSS: For styling and enhancing the Streamlit interface

Implementation Details

The application is built with a Streamlit frontend that provides an intuitive interface for users to input story prompts and adjust generation parameters. On the backend, it communicates with Ollama, which runs local LLM models like 'phi' or other compatible models.

When a user submits a prompt, the application processes it through the selected LLM model, which generates a creative story based on the input. The story is then displayed in real-time on the interface, allowing for immediate feedback and iteration.

Future Enhancements

• Support for multiple LLM models with option to select preferred model
• Story formatting and export options (PDF, DOCX, etc.)
• Collaborative storytelling features
• Speech-to-text and text-to-speech capabilities
• Integration with image generation for illustrated stories