aotakeda/learn-thing: Create mind maps to learn new things using AI.

This is a simple Next.js project that implements a mind map visualization tool using React Flow.

Watch a demo of it in action here.

The UI is built using shadcn and some components from Magic UI.

It allows users to view and interact with mind maps, and download the mind map data as a markdown file.

The mind map data is generated using either local models from Ollama or external models like OpenAI and leveraging AI SDK.

• Interactive mind map visualization
• Node details view in a side sheet
• Markdown export functionality
• Save mind map data to a local JSON file
• Switch between local and external models

Install all dependencies:

Copy the .env.template file to .env.local and specify which model (local or external) you want to use by setting the NEXT_PUBLIC_USE_LOCAL_MODELS environment variable to true or false.

When running an OpenAI model, you must specify your OpenAI API key in the .env.local file.

Inside the route.ts file, you must specify the model you are running using Ollama, by default it will use the llama3.1 model for local models and for external models it will use the gpt-3.5-turbo model.

Bear in mind that external models tend to be much faster serving than local models.

If you want to learn how to run a model locally, check out the Ollama documentation.

Now you’re ready to run the development server:

Open http://localhost:3000 with your browser and then start creating your own learning mind maps.

The prompts used to generate the mind map data is defined in the defaultLocalPrompt and defaultExternalPrompt variables in the prompts.ts file.

This project is licensed under the MIT License.