Presenton Local Docker Deployment Demo
by Trevor Lee in Design > Software
14 Views, 0 Favorites, 0 Comments
Presenton Local Docker Deployment Demo
This simple demo is targeted to showcase the local deployment of the LLM-powered presentation slides generation tool Presenton -- Open-Source AI Presentation Generator and API (Gamma, Beautiful AI, Decktopus Alternative)
Presenton is an open-source application for generating presentations with AI — all running locally on your device. Stay in control of your data and privacy while using models like OpenAI and Gemini, or use your own hosted models through Ollama.
The demo is targeted to show audience:
- Deployment of Presenton locally using Docker, specifically Docker Desktop.
If you have Docker Desktop installed on your computer, I guess you are as technical savvy as to follow the steps shown here.
- Hooking up Presenton with OpenRouter LLM gateway for using selected model on a pay-as-you-go basis.
Yes, you will have to pay for the LLM service you use. However, it likely will not cost you much unless you are a very very heavy user.
- Hooking up Presenton with Pixabay API for image [auto] sourcing.
It is a free service.
OpenRouter Account Setup
To setup an OpenRouter account, head to OpenRouter.ai. If not yet, sign up for an account. You can use your Google account for signing up.
Then, the first thing you might want to do is to check out OpenRouter's pricing / crediting system
- You pay for what you use with credits -- US$ -- you pre-purchased.
- Different models have different crediting rates. The more powerful the model, the higher the rate. For example, as of 2026-02-01, DeepSeek V3.2 will cost $0.25 per a million input tokens and $0.38 per a million output tokens. Simply put, roughly, you paid $0.25 if your input prompt has 1 million words, and you paid $0.38 if the output generated by the model has 1 million words. Note that there are some free models you can choose from; e.g. DeepSeek: R1 0528 is free.
Since you will be using OpenRouter as the LLM service provider for Presenton, you need to get your API key from OpenRouter first. As will be shown later, you will need this key to deploy Presenton with Docker.
Once created, you will be assigned a new API key. When the API key is used, OpenRouter will automatically match the API key with any restriction imposed on the use of that API key, like Credit limit. Anyway, copy the key somewhere safe for later use.
Once the page is closed, there is no way for you to find out the key again. You will have to generate a new key if you lost it.
Pixabay Account Setup
You also need to setup a free account with Pixabay to get your API key for image sourcing. Again, you can use your Google account for signing up.
To find out the API key assigned to your account, go to their API documentation page (as shown in above screenshot).
To setup Presenton with Docker, you will need this API key later.
Demo of Slides Generation With Presenton
Before getting into the deployment, here I will show you a demo session of slide generation first. Hopefully, the demo can give you a sense of what kind of result you can get; more importantly, a feel of how much you will pay the result you get.
The prompt I am going to use for the slides generation is:
Notice that I have chosen 5 slides in English (above screenshot).
Even though I can supply additional Supporting Documents, I will just skip the option.
The first thing it does is to generate the slide outline, which you can double check (as you should be) and edit whenever necessary.
Next, you will need to choose a template for the slides (as of current version, there are not many templates to choose from 😔)
I will choose the first in-built one (see above screenshot)
After a while, all the 5 slides are generated from the outline. (Notice from a screenshot above of the Speaker notes.)
Presenton also provides some basic editing features for you to modify the generated slides.
For example, you can change the image on the slide to another one. (As can be seen from a screenshot above, even though the option provided is AI Generate, it doesn't seem to ask the OpenRouter model to generate new images. Instead, it seems that it just sources other images from Pixabay based on the Image Description you entered. Another alternative is to upload your own image from your local computer.)
Of course, you can export the slides in PPTX (PowerPoint) format, which you can further edit locally with LibreOffice Impress (Microsoft PowerPoint free alternative), or upload to Google Slides for online editing, sharing etc.
Note that the created / generated slides are stored locally in your Docker storage. And you can go back to the slides from the Dashboard.
How much did it cost me to generate these 5 slides? Basically, close to nothing. According to the OpenRouter usage records, the slide generation called DeepSeek V3.2 7 times. And the 7 LLM calls costed me around 0.00445 USD in total.
Local Deployment of Presenton With Docker
As mentioned previously, Presenton can be deployed locally to your Docker Desktop installation. What you need to do is
- Create a folder, say local_presenton
- In the folder, you will need a Docker configuration file -- a docker-compose.yml file -- you can refer to the docker-compose.yml provided in this GitHub repo
- You will also need to create a .env file in the same folder to tell Docker your secrets, like your OpenRouter and Pixabay API keys. You can refer to the sample .env.example provided in this GitHub repo
You can change the model to use by changing OPENAI_MODEL. E.g. you can change to use qwen3-30b-a3b-instruct-2507 like
The list of models you can choose from, as well as their crediting rates, can be found at OpenRouter Models
Open a Command Prompt / PowerShell Terminal, change directory to the local_presenton folder you created, and run the following command to start the Presenton Docker service locally
This will deploy Presenton locally as a Docker container.
As can be seen from on of the Docker Desktop containers screenshot above:
- local_presenton -- the folder name -- is the name of your Docker container [group]
- presenton_demo is the name of the Presenton Docker container
Finally, to enjoy your local Presenton deployment (a web app), in your browser, go to http://localhost:8080
Enjoy!
Hope this little demo can help you. Have fun!
Peace be with you! May God bless you! Jesus loves you! Amazing Grace!