Presenton

What is Presenton?

Presenton is an open source AI presentation generator that turns a prompt, an uploaded document, or structured data into an editable slide deck. It runs as a self-hosted Docker service, a desktop app, or a managed cloud, and exports fully editable PowerPoint (PPTX) and PDF files. You bring your own AI models, so your content and API keys stay under your control.

What is Presenton best for?

Presenton fits privacy-sensitive teams and developers who want deck generation without routing content through a proprietary SaaS. Its REST API and MCP endpoint make it a strong pick for embedding presentation generation into a product or automating recurring reports, while the desktop and Docker options suit individuals who just want local, model-agnostic slide creation.

What can Presenton do?

• Generate decks from a text prompt, or from uploaded PDF, DOCX, and other documents
• Export fully editable PPTX and clean PDF files, plus a drag-and-drop canvas editor
• Bring your own models across OpenAI, Anthropic Claude, Google Gemini, Vertex AI, Azure OpenAI, Amazon Bedrock, DeepSeek, OpenRouter, and any OpenAI-compatible endpoint
• Run models locally through Ollama or LM Studio for full-privacy, offline generation
• Import an existing PPTX to establish brand templates, or build custom templates in HTML and Tailwind CSS
• Automate generation over a REST API (/api/v1/ppt/presentation/generate) and an MCP endpoint, with async workflows and scoped iframe embed tokens
• Generate slide images via DALL-E 3, Gemini, Pexels, Pixabay, or ComfyUI

Is Presenton free?

Yes — the self-hosted version is free and open source under the Apache-2.0 license; you only pay for your own server and whatever AI provider you plug in (or nothing, if you run local models). Presenton Cloud is a separate paid, credit-based service (one slide equals one credit) for teams that want a managed platform with premium templates and collaboration.

Where does Presenton fall short?

• Output quality and cost depend entirely on the AI provider you connect — there’s no bundled model, so you supply keys (or GPU capacity for local models) and results vary by model.
• Image generation and web-search grounding need external provider keys or extra services, so a fully local, private setup gives up those features unless you add ComfyUI and similar tooling.
• The self-hosted stack spans a Next.js frontend and a Python FastAPI backend (Node.js, Python 3.11, and uv for a from-source setup), which is more moving parts than a single-binary tool, and the MCP endpoint isn’t available in the Electron desktop app.

What does Presenton replace?

Presenton is a self-hosted alternative to Gamma, Beautiful.ai, and the presentation side of Canva. It does the same prompt-to-deck job, but you run it on your own infrastructure, choose your own AI models, and aren’t locked into per-credit or per-seat SaaS pricing.

FAQ

Is Presenton open source? Yes. Presenton is released under the Apache-2.0 license, a permissive OSI-approved license, and the full source is on GitHub.

Can I self-host Presenton for free? Yes. Self-hosting is free — you only pay for the server and any AI provider you use. Point it at a local model through Ollama or LM Studio and you can run it at no per-slide cost, unlike the paid Presenton Cloud.

Is Presenton a good Gamma alternative? For teams that want data control, an API, and model choice, yes — it generates editable PPTX and PDF decks and lets you bring your own models. Gamma may still feel more polished out of the box, since Presenton’s quality tracks the model you connect.

What do I need to run Presenton? The simplest path is Docker (with optional GPU support). A from-source install needs Node.js, Python 3.11, and uv, plus an API key for a hosted AI provider or a local model runner such as Ollama.