The file-system chaos you're fed up with—solved

Every computer user knows the feeling: a sprawling Downloads folder, a desktop littered with PDFs, screenshots, and random zip files, and the constant mental overhead of trying to remember where you saved that one important receipt. Traditional file managers give you a view, but they don't understand what's inside the files, nor do they help you keep the structure tidy as you work.

Enter LlamaFS. Built on top of Llama 3, it watches your filesystem, learns from your actions, and instantly proposes (or applies) a sensible hierarchy—renaming files, creating date-based folders, and even extracting semantic tags from images and audio. All of this happens in under 500 ms per operation, thanks to smart caching and a lightweight Python + Electron stack.

Why Choose LlamaFS?

• AI-driven organization – Llama 3 analyzes file content (text, images via Moondream, audio via Whisper) to suggest meaningful names and folder structures.
• Two operating modes – Batch mode lets you run a one-off organization pass on any directory; Watch mode runs as a daemon, continuously learning from your edits.
• Privacy-first incognito mode – Route every LLM request through a local Ollama instance, keeping your personal data off the cloud.
• Lightning-fast feedback – Most file operations complete in under 500 ms, making the experience feel instantaneous.
• Cross-platform UI – An Electron front-end gives a polished, native-looking interface on Windows, macOS, and Linux.
• Open-source freedom – No vendor lock-in, no subscription fees, and the ability to fork or extend the codebase as you wish.

Spotlight on Key Features

Getting Started

Below is a quick path from zero to a running LlamaFS instance on any modern OS.

Prerequisites

• Python 3.10+ (for the backend)
• Node.js 20+ (for the Electron UI)
• Optional but recommended: Ollama if you want incognito mode.

Installation Steps

# 1. Clone the repository
git clone https://github.com/iyaja/llama-fs.git
cd llama-fs

# 2. Set up the Python environment
python -m venv .venv
source .venv/bin/activate   # on Windows: .venv\Scripts\activate
pip install -r requirements.txt

# 3. Install the Electron front-end
npm install
npm run build   # compiles the UI

# 4. Run LlamaFS in your preferred mode
# Batch mode – organize a folder once
python src/main.py --mode batch --path ~/Downloads

# Watch mode – start the daemon
python src/main.py --mode watch --path ~/Desktop

Tip: To enable incognito mode, start Ollama locally (ollama serve) and launch LlamaFS with --incognito. All LLM calls will be routed to the local model, guaranteeing zero data leaves your device.

How LlamaFS Stacks Up Against Proprietary Alternatives

Bottom line: If you need true AI-powered organization without surrendering your data to a SaaS vendor, LlamaFS is the only open-source option that offers both cloud and on-premise LLM inference.

Call to Action

Ready to turn your chaotic file system into a smart, self-organizing library?

• Try the project: https://github.com/iyaja/llama-fs
• Star the repo: https://github.com/iyaja/llama-fs
• Read the docs: Getting Started Guide
• Join the community: Open an issue or pull request on GitHub to suggest new features, report bugs, or share your custom sorting rules.

LlamaFS proves that you don't need a pricey commercial tool to get intelligent file management. With zero vendor lock-in, full privacy control, and a vibrant open-source community, it's the future-ready solution for anyone who wants their files to work for them—not the other way around.