Private local AI personal knowledge management app.

Announcement

We are now on Discord! Our team is shipping very quickly right now, so sharing feedback with us will really help shape the product.

About

Reor is an AI-powered desktop note-taking app designed for seamless knowledge management. With features that automatically link related notes, answer questions, provide semantic searches, and generate AI flashcards, Reor is tailored for efficiency. What sets Reor apart is that everything is stored locally, allowing users to edit notes with an Obsidian-like markdown editor. This project is built on the foundation that AI tools for thought should prioritize local model operations.

Reor utilizes advancements from [Ollama](https://github.com/ollama/ollama), [Transformers.js](https://github.com/xenova/transformers.js), and [LanceDB](https://github.com/lancedb/lancedb) to enable local execution of both LLMs and embedding models:

• Every note you write is chunked and embedded into an internal vector database.
• Related notes are connected automatically via vector similarity.
• LLM-powered QA performs RAG on your corpus of notes.
• All content can be searched semantically.

Imagine Reor as a library with two librarians: one is the LLM, accessing a database of information to answer your questions, and the other is you, bringing your current thoughts to connect with ideas already available in the library. When you inquire about a topic, the LLM retrieves relevant information from your notes to provide answers, while you can also explore related notes to expand your understanding and insights.

Getting Started

1. Download from reorproject.org or releases. Mac, Linux, and Windows are all supported.
2. Install the app like any regular application.

Running Local Models

Reor interacts directly with Ollama, enabling you to download and run models locally. To do this, navigate to Settings – Add New Local LLM and enter the name of the model you wish to download. You can find available models here. You may also connect to an OpenAI-compatible API like Oobabooga, Ollama, or OpenAI itself.

Importing Notes from Other Apps

Reor operates within a specified directory. Choose your directory during the initial boot. To import notes from other applications, you will need to manually populate that directory with markdown files. Be mindful that front matter in your markdown files may not parse correctly. We’re working on integrations with other apps!

Building from Source

Ensure you have nodejs installed.

Follow these steps:

1. Clone the repository:

git clone https://github.com/reorproject/reor.git

2. Install dependencies:

npm install

3. Run for development:

npm run dev

4. Build:

npm run build

Interested in Contributing?

We always welcome contributors who are passionate about enhancing the future of knowledge management. Whether you have a feature idea, want to squash a bug, or seek to improve styling, we’d love to hear from you. Check out our issues page and the contributing guide to get started.

Troubleshooting

If you encounter any issues while using Reor, consider the following troubleshooting tips:

• Ensure all dependencies are correctly installed.
• Check your internet connection, especially when downloading models.
• If your notes are not importing correctly, verify that your markdown files are properly formatted.
• For any bugs, check the issues page on our GitHub Repo.

For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.

At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.