LandPPT: AI platform for automatic generation of PPT presentations

LandPPT is a Large Language Model (LLM) based presentation generation platform that automatically converts user-supplied document content into professionally designed PPT presentations. The core role of this platform is to simplify the PPT production process, where users only need to enter a topic or upload a document, and the AI generates an outline and ultimately creates a complete presentation. It accesses a variety of mainstream AI models, such as OpenAI's GPT series, Anthropic's Claude series, and Google's Gemini series, and also supports open source models deployed locally through Ollama. In addition to AI-driven content generation, LandPPT also provides a wealth of template and style options, allowing users to easily create high-quality presentations that meet professional requirements, and supports exporting to a variety of formats such as PDF, HTML, and PPTX to suit different presentation scenarios.

Function List

• Support for multiple AI models: The platform integrates OpenAI, Anthropic, Google, Azure OpenAI, and localized Ollama models, allowing users to choose the most appropriate AI for content generation according to their needs.
• Powerful document processing capabilities: Support for direct processing of PDF, Word, Markdown and TXT and other formats of the document, and use theminuerucap (a poem)markitdownand other tools to intelligently extract core content.
• In-depth research function: Integration with the Tavily API allows for networking depth research based on user-entered topics, ensuring depth and breadth of presentation content.
• Rich template system: Provides unified global master templates, diverse creative page layouts generated by AI, and also supports users to import and create customized personalized templates.
• Complete project management process: The workflow is divided into three stages: requirements confirmation, outline generation and PPT generation, with a clear process. Users can also modify and confirm the AI-generated outline through the visual editor.
• Modernized web interface: Provides an intuitive and easy-to-use web interface that supports 16:9 ratio responsive page preview in real time. Users can also make real-time edits and adjustments to the presentation content through the AI chat function in the sidebar.
• Supports multiple export formats: Users can export their finalized presentations as PDF, HTML or PPTX files for easy use on different devices and platforms.

Using Help

LandPPT, as a powerful AI presentation generation tool, provides two ways to use it, local installation and Docker deployment, which users can choose according to their technical background and needs.

System Environment Requirements

Before you begin, make sure your system meets the following basic requirements:

• Python::3.11or later
• comprehensive database: SQLite 3
• Docker: (Optional) If you wish to use containerized deployment, you need to install Docker

Method 1: Local installation (uv recommended)

For users familiar with the Python environment, local installation is the most straightforward way. The project recommends usinguvThis is a very fast Python package installation and management tool.

Step 1: Clone the project code
Open a terminal and use thegitcommand clones the LandPPT source code from GitHub to your local computer.

git clone https://github.com/sligter/LandPPT.git

Then go to the project directory.

cd LandPPT```
**第二步：安装uv**
如果您的系统中还没有安装`uv`，可以根据您的操作系统选择对应的命令进行安装。
- **Windows (使用PowerShell)**:
```powershell
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

• macOS or Linux:

  curl -LsSf https://astral.sh/uv/install.sh | sh
  

Step 3: Install project dependencies
Project dependencies are recorded in thepyproject.tomlfile. Use theuv(used form a nominal expression)synccommand to quickly create a virtual environment and install all necessary packages.

uv sync

In order to support export in PPTX format, the Apryse SDK also needs to be installed.

uv pip install apryse-sdk --extra-index-url=https://pypi.apryse.com

Step 4: Configure Environment Variables
The project needs to be configured with sensitive information such as API keys. You can do this from the template file.env.exampleMake a new copy of the configuration file.

cp .env.example .env```
接下来，使用文本编辑器打开`.env`文件，填入您自己的AI API密钥。例如，如果您计划使用OpenAI，就需要填写`OPENAI_API_KEY`。如果需要联网研究功能，则需要填写`TAVILY_API_KEY`。
```ini
# AI提供商配置
DEFAULT_AI_PROVIDER=openai
OPENAI_API_KEY=your_openai_api_key_here
ANTHROPIC_API_KEY=your_anthropic_api_key_here
GOOGLE_API_KEY=your_google_api_key_here
# 功能配置
TAVILY_API_KEY=your_tavily_api_key_here
APRYSE_LICENSE_KEY=your_apryse_key_here # 用于PPTX导出

Step 5: Start the service
After completing all the configurations, run the following command to start the web service.

uv run python run.py

Once the service has started, you will be able to access thehttp://localhost:8000to use LandPPT now.

Method 2: Deployment using Docker

For users who want to simplify the configuration of their environment, using Docker is the best option.

Step 1: Pull the pre-built Docker image
The project developer has already packaged the application as a Docker image and uploaded it to Docker Hub, and you can pull the latest image with a single command.

docker pull bradleylzh/landppt:latest

Step 2: Prepare the configuration file
As with a local installation, you still need to create a hosted.envfile and fill in the necessary API keys.

Step 3: Run the container
utilizationdocker runcommand to start the LandPPT container. This command starts the LandPPT container with your local.envThe files are mounted inside the container and the data persistence volume is set to ensure that your project data is not lost.

docker run -d \
--name landppt \
-p 8000:8000 \
-v $(pwd)/.env:/app/.env \
-v landppt_data:/app/data \
-v landppt_reports:/app/research_reports \
-v landppt_cache:/app/temp \
bradleylzh/landppt:latest

Step 4: Access and management
Once the container is successfully running, again via a browser, access thehttp://localhost:8000The operation interface can be opened. If you need to view the log of the service running, you can use the following command:

docker logs -f landppt

workflow

1. Accessing the Web Interface: After starting the service, type in your browserhttp://localhost:8000The
2. Configuring AI: In the Settings page of the interface, fill in the API key for the AI model you are using.
3. Create a project: On the main screen, enter the topic of the presentation you want to generate, select the audience (e.g. "technical staff" or "marketing team") and the desired range of PPT pages.
4. Generating an Outline: Click the Generate button and AI will automatically generate a well-structured presentation outline according to your needs. You can modify and confirm the content in the outline editor.
5. Generate PPT: After confirming that the outline is correct, AI will generate a complete HTML presentation with real-time preview based on the outline content and built-in templates.
6. Editing and exporting: Utilizing the AI Chat feature in the sidebar, you can fine-tune the generated content. After finishing all edits, you can export the presentation to PDF, HTML or PPTX format.

application scenario

1. Business Analyst
   Business analysts need to report regularly to management on market trends, competitive product analysis or project progress. They can use LandPPT to quickly generate clearly structured and illustrated presentations by uploading data reports or typing in analysis topics, so they can devote more time to data analysis itself rather than PPT layout.
2. Educators and students
   Teachers can use the tool to quickly convert course handouts (e.g. Word or PDF documents) into vivid teaching courseware. Students can also quickly generate the first draft of a presentation by entering a thesis outline or a research paper when preparing a course report or a graduation design defense, and then make personalized changes, which greatly enhances the efficiency of learning and research.
3. marketer
   Marketing teams need to create a large number of presentations when developing marketing campaign programs or making product introductions. With LandPPT, they can input campaign themes and core ideas and let AI automatically generate presentations containing modules such as market positioning, target users, promotion strategies, etc., so as to quickly respond to market changes.
4. Technology Developer
   Developers can use LandPPT to convert technical documents or Markdown notes directly into presentations when they are sharing technology, reviewing projects or reporting technical solutions to superiors. This not only saves production time, but also makes the presentation of technical content more systematic and professional.

QA

1. What language models does LandPPT support? Can I use my own local model?
   The LandPPT platform supports a wide range of mainstream cloud-based large-scale language models, including OpenAI's GPT series, Anthropic's Claude series, Google's Gemini series, and Microsoft's Azure OpenAI service. It also supports accessing and using your own locally deployed open source models through the Ollama framework, providing users with great flexibility.
2. What do I do if the AI-generated outline doesn't meet my requirements?
   Outline generation is a separate part of the LandPPT workflow; after AI generates the outline, you can access a visual "outline editor" where you can manually modify, add or delete any part of the content. Only after you have verified that the outline is correct will the platform move on to the next step of PPT generation to ensure that the final content meets your expectations.
3. Can I customize the templates and styles of my presentation?
   Yes. LandPPT has a powerful template system. In addition to the global master templates provided by the platform and the diverse layouts generated by AI, it also supports users to import and create their own personalized templates. This means that you can design the look of your presentations according to your company's brand guidelines or personal preferences for a high degree of customization.
4. Do I need to pay to export my project file to PPTX format?
   LandPPT itself is an open source project and most of its features are free. However, to enable exporting HTML presentations to PPTX format, the project integrates the Apryse SDK, and according to Apryse's licensing policy, you may need to obtain the appropriate license key (APRYSE_LICENSE_KEY) before you can use this export feature in your production environment.