PromptMage is a Python framework designed to simplify the development of complex, multi-step applications based on Large Language Models (LLMs). As a self-hosted solution, it offers an intuitive interface that streamlines the creation and management of LLM workflows. PromptMage facilitates prompt testing and comparison, incorporating version control features to help users track the development of their prompts. Suitable for both small teams and large enterprises, PromptMage aims to enhance productivity and promote the practical use of LLM technology. Key Features and Functionality: - Prompt Playground: An interactive interface for testing, comparing, and refining prompts, enabling rapid iteration and development. - Integrated Version Control: Built-in tracking system that allows seamless collaboration and iteration by monitoring prompt development. - Auto-Generated API: Utilizes FastAPI to automatically create APIs, facilitating easy integration and deployment of LLM applications. - Evaluation Mode: Supports both manual and automated testing to assess prompt performance, ensuring reliability before deployment. - Type Hinting: Comprehensive type hinting for automatic inference and validation, streamlining the development process. Primary Value and Problem Solved: PromptMage addresses the challenges associated with developing and managing complex LLM workflows by providing a cohesive platform that integrates prompt testing, version control, and deployment. By offering an intuitive interface and robust features, it enhances productivity, fosters collaboration, and supports the practical application of LLM technology across various organizational scales.