presenton/presenton

Presenton – Open‑Source AI‑Powered Presentation Builder

What it is – Presenton is a self‑hosted web/desktop application that lets you create PowerPoint‑style slide decks automatically with large language models. You can feed it a plain‑text prompt, upload a document, or start from one of the built‑in slide‑template collections, and the system will generate fully editable .pptx (and PDF) files. The core is an LLM‑driven generation engine plus a drag‑and‑drop slide editor.

Key capabilities

• Model‑agnostic: works with any OpenAI‑compatible provider (OpenAI, Anthropic, Gemini, Azure, Bedrock, Ollama, LM Studio, etc.) or local models via Ollama.
• Multiple deployment options: one‑command Docker container, a cross‑platform Electron desktop app (macOS, Windows, Linux), or a cloud‑hosted SaaS version.
• Template library: ready‑made business, pitch‑deck, and educational slide templates that can be previewed and customized.
• Editable output: generated decks are exported as fully editable PowerPoint files, not just images or PDFs.
• API: an HTTP API (Model Context Protocol) lets you integrate the generator into other tools or run it as a private service for a team.
• Enterprise features: Helm charts for Kubernetes, SSO/OIDC/SCIM, audit logs, and admin console for multi‑user workspaces.
• Privacy‑first: you control the model and API keys; all processing can run locally with no cloud‑side data collection.

How to get started

1. Docker – docker run -it --name presenton -p 5001:80 -v "$(pwd)/app_data:/app_data" ghcr.io/presenton/presenton:latest then open http://localhost:5001.
2. Desktop – download the appropriate installer (.dmg, .exe, .deb) from the website and run the Electron app. The first‑time setup creates a .env file where you place your LLM provider keys.
3. Configuration – set environment variables such as LLM=openai, OPENAI_API_KEY, OPENAI_MODEL=gpt-4.1, or choose LLM=ollama for a local model. Additional variables control generation mode, community gallery, and key‑visibility.

Typical workflow

• Choose a template (e.g., Momentum for a business report).
• Enter a short prompt like “Create a quarterly sales deck for Q3 2024” or upload a PDF/Word file.
• The AI fills in slide titles, bullet points, charts, and images.
• Refine the deck with the built‑in drag‑and‑drop editor.
• Export the final deck as .pptx or PDF, or call the API to generate slides programmatically.

Why it matters Presenton bridges the gap between generic LLM chat tools and specialized presentation software. By being open‑source and self‑hostable, it avoids SaaS lock‑in, lets organizations keep proprietary content on‑prem, and supports a wide range of LLM back‑ends—including free, locally‑run models.

---

Quick links

• Docs & quick‑start: https://docs.presenton.ai/
• Download desktop: https://presenton.ai/download
• Docker image: ghcr.io/presenton/presenton
• Community templates: https://presenton.ai/explore

---

Bottom line – If you need AI‑generated slide decks without giving up control of your data or paying recurring SaaS fees, Presenton provides a full‑stack, extensible solution that you can run anywhere—from a local laptop to a Kubernetes cluster.