In today's post, we’re diving into one of the easiest and most secure ways to generate powerful slide presentations using AI—all from your local machine. Whether you're a content creator, educator, or business professional, Presenton lets you generate customizable decks from simple prompts or uploaded documents in seconds.
Let’s walk through how it works, its features, and how you can get started using either OpenAI models or fully offline, local open-source models with Ollama.
What is Presenton?
Presenton is an open-source AI presentation generator that runs on your local device via Docker. It allows you to create high-quality presentations by simply entering a prompt or uploading a document (PDF, DOCX, TXT, PPTX, etc.).
✅ Key Benefits:
- Fully Local Execution: No internet required for generation (when using local models).
- Privacy First: No data tracking or storage. Your files and prompts stay with you.
- Supports Multiple Models: Use OpenAI, Gemini, or local models via Ollama.
- Image Integration: Automatically adds relevant visuals using providers like DALL·E 3, Gemini Flash, Pexels, and Pixabay.
- Flexible Export: Export your slides in PPTX or PDF format.
- Open-Source Freedom: Licensed under Apache 2.0, free to modify or extend.
🔧 Prerequisites and Setup
To run Presenton locally, you’ll need:
- Docker installed (you can use the VS Code Docker extension).
- Python environment (optional, for easier management).
- Access to API keys (if using OpenAI, Gemini, Pexels, etc.).
Installation Steps (Simplified):
-
Install Docker and verify it using:
docker --version/Docker AI EAP : How to Get Started | Docker -
Clone the Presenton Repository:
git clone https://github.com/presenton/presenton.git cd presenton
-
Run this command in Windows "docker run -it --name presenton -p 5000:80 -v "${PWD}\app_data:/app_data" ghcr.io/presenton/presenton:latest" -
Access the app at http://localhost:5000
-
Paste your OpenAI API Key or choose Ollama for offline generation.
💡 Creating a Presentation – Step-by-Step
You can generate a slide deck in seconds. Here’s how:
Method 1: Using OpenAI (Online)
-
Select your AI provider (e.g., OpenAI).
-
Enter your API key.
-
Type a prompt, e.g., “How to lose weight without medication”.
-
Choose number of slides and your language.
-
Select a layout (classic, modern, general).
-
Click “Generate.”
✅ Output: A fully outlined, image-enhanced presentation based on your prompt. You can edit text, change the title, and export your slides.
Method 2: Using Ollama (Offline with Local Models)
Prefer offline and private? Ollama is your go-to.
-
Download and install Ollama.
-
Load a local model like
llama3:8b:ollama run llama3 -
In Presenton, choose "Ollama" as your model backend.
-
Type a prompt like “How to pass any job interview”.
-
Select the number of slides and click “Generate.”
Even the smaller LLaMA 3 models provide great results, complete with content-rich slides and visuals.
🖼️ Visual Support
Presenton integrates with popular image APIs to enhance your slides:
-
DALL·E 3
-
Gemini Flash
-
Pexels (get API key here)
-
Pixabay
Just enter your API key, and Presenton will auto-generate relevant images based on your content.
🧪 Tested and Approved
I tested this tool with both online and offline models. Results were impressive:
-
Speed: Generated slide decks in under a minute.
-
Accuracy: Topics were well-structured and logically outlined.
-
Customization: Easy to modify layouts, text, and images.
-
Security: Running everything locally ensured full control over data.
🔚 Final Thoughts
If you're looking for a powerful, flexible, and privacy-focused way to create presentations with AI—Presenton is worth trying. With support for multiple models and formats, it's the ultimate tool for educators, professionals, and creators who care about both productivity and data control.
🔗 Resources and Links
-
👉 Presenton GitHub Repository (Insert actual repo link)
📺 Watch the Full Tutorial
🎥 Don’t forget to check out the full walkthrough video on YouTube. Like, comment, and subscribe for more hands-on AI tools and tutorials!