Presenton is an open-source AI-powered presentation generator designed for users who want to create presentations locally, maintaining data privacy. It offers an alternative to cloud-based services like Gamma and Beautiful AI, supporting various LLMs and image generation models.

How It Works

Presenton utilizes a modular architecture with a FastAPI backend and Next.js frontend, orchestrated by a Model Context Protocol (MCP) server. This MCP server manages a state machine for the presentation generation workflow, including session management, outline generation, content and asset creation using user-selected AI models, and export to PPTX or PDF. This approach provides flexibility and extensibility for customization.

Quick Start & Requirements

• Install/Run: Via Docker: docker run -it --name presenton -p 5000:80 -v "./app_data:/app_data" ghcr.io/presenton/presenton:latest
• Prerequisites: Docker. For GPU support with Ollama, NVIDIA Container Toolkit is required.
• Configuration: API keys for LLMs (OpenAI, Google Gemini, Anthropic Claude) and image providers (DALL-E 3, Pexels, Pixabay) can be set via environment variables.
• Access: Open http://localhost:5000 in a browser.
• Docs: Official Documentation

Highlighted Details

• Supports local model execution via Ollama for enhanced privacy.
• Offers API endpoints for programmatic presentation generation.
• Allows custom layouts using HTML and Tailwind CSS.
• Integrates with VS Code for workflow management via chat commands.

Maintenance & Community

• Community: Discord server available.
• License: Apache 2.0.

Licensing & Compatibility

Apache 2.0 license permits commercial use and modification, with attribution.

Limitations & Caveats

The project is actively developed, with some features marked as planned (e.g., user-configurable system prompts). GPU acceleration is primarily noted for Ollama models.