[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-73559":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":27,"readmeContent":28,"aiSummary":29,"trendingCount":16,"starSnapshotCount":16,"syncStatus":18,"lastSyncTime":30,"discoverSource":31},73559,"llama-fs","iyaja\u002Fllama-fs","iyaja","A self-organizing file system with llama 3","",null,"TypeScript",5732,378,45,53,0,1,2,5,3,38.74,"MIT License",false,"main",true,[],"2026-06-12 02:03:15","# LlamaFS\n\n\u003Cimg src=\"electron-react-app\u002Fassets\u002Fllama_fs.png\" width=\"30%\" \u002F>\n\n## Inspiration\n\n[Watch the explainer video](https:\u002F\u002Fx.com\u002FAlexReibman\u002Fstatus\u002F1789895425828204553)\n\nOpen your `~\u002FDownloads` directory. Or your Desktop. It's probably a mess...\n\n> There are only two hard things in Computer Science: cache invalidation and **naming things**.\n\n## What it does\n\nLlamaFS is a self-organizing file manager. It automatically renames and organizes your files based on their content and well-known conventions (e.g., time). It supports many kinds of files, including images (through Moondream) and audio (through Whisper).\n\nLlamaFS runs in two \"modes\" - as a batch job (batch mode), and an interactive daemon (watch mode).\n\nIn batch mode, you can send a directory to LlamaFS, and it will return a suggested file structure and organize your files.\n\nIn watch mode, LlamaFS starts a daemon that watches your directory. It intercepts all filesystem operations and uses your most recent edits to proactively learn how you rename file. For example, if you create a folder for your 2023 tax documents, and start moving 1-3 files in it, LlamaFS will automatically create and move the files for you!\n\nUh... Sending all my personal files to an API provider?! No thank you!\n\nIt also has a toggle for \"incognito mode,\" allowing you route every request through Ollama instead of Groq. Since they use the same Llama 3 model, the perform identically.\n\n## How we built it\n\nWe built LlamaFS on a Python backend, leveraging the Llama3 model through Groq for file content summarization and tree structuring. For local processing, we integrated Ollama running the same model to ensure privacy in incognito mode. The frontend is crafted with Electron, providing a sleek, user-friendly interface that allows users to interact with the suggested file structures before finalizing changes.\n\n- **It's extremely fast!** (by LLM standards)! Most file operations are processed in \u003C500ms in watch mode (benchmarked by [AgentOps](https:\u002F\u002Fagentops.ai\u002F?utm_source=llama-fs)). This is because of our smart caching that selectively rewrites sections of the index based on the minimum necessary filesystem diff. And of course, Groq's super fast inference API. 😉\n\n- **It's immediately useful** - It's very low friction to use and addresses a problem almost everyone has. We started using it ourselves on this project (very Meta).\n\n## What's next for LlamaFS\n\n- Find and remove old\u002Funused files\n- We have some really cool ideas for - filesystem diffs are hard...\n\n## Installation\n\n### Prerequisites\n\nBefore installing, ensure you have the following requirements:\n- Python 3.10 or higher\n- pip (Python package installer)\n\n### Installing\n\nTo install the project, follow these steps:\n1. Clone the repository:\n   ```bash\n   git clone https:\u002F\u002Fgithub.com\u002Fiyaja\u002Fllama-fs.git\n   ```\n\n2. Navigate to the project directory:\n    ```bash\n    cd llama-fs\n    ```\n\n3. Install requirements\n   ```bash\n   pip install -r requirements.txt\n   ```\n\n4. Update your `.env`\nCopy `.env.example` into a new file called `.env`. Then, provide the following API keys:\n* Groq: You can obtain one from [here](https:\u002F\u002Fconsole.groq.com\u002Fkeys).\n* AgentOps: You can obtain one from [here](https:\u002F\u002Fapp.agentops.ai\u002Fsettings\u002Fprojects).\n\nGroq is used for fast cloud inference but can be replaced with Ollama in the code directly (TODO.)\n\nAgentOps is used for logging and monitoring and will report the latency, cost per session, and give you a full session replay of each LlamaFS call.\n\n5. (Optional) Install moondream if you want to use the incognito mode\n    ```bash\n    ollama pull moondream\n    ```\n\n## Usage\n\nTo serve the application locally using FastAPI, run the following command\n   ```bash\n   fastapi dev server.py\n   ```\n\nThis will run the server by default on port 8000. The API can be queried using a `curl` command, and passing in the file path as the argument. For example, on the Downloads folder:\n   ```bash\n   curl -X POST http:\u002F\u002F127.0.0.1:8000\u002Fbatch \\\n    -H \"Content-Type: application\u002Fjson\" \\\n    -d '{\"path\": \"\u002FUsers\u002F\u003Cusername>\u002FDownloads\u002F\", \"instruction\": \"string\", \"incognito\": false}'\n   ```\n","LlamaFS 是一个自组织文件管理系统，能够根据文件内容和常见规范（如时间）自动重命名和整理文件。其核心功能包括支持多种文件类型（如通过Moondream处理的图像和通过Whisper处理的音频），并提供两种运行模式：批处理模式和监控模式。在批处理模式下，用户可以将整个目录提交给LlamaFS以获得建议的文件结构；而在监控模式下，LlamaFS会启动一个守护进程来观察指定目录，学习用户的文件管理习惯并主动执行文件移动等操作。此外，该系统还提供了隐私保护选项“隐身模式”，确保所有请求仅通过本地Ollama处理而不暴露给外部API提供商。LlamaFS非常适合需要高效管理和组织个人或团队文件夹的场景，特别是那些经常面临文件混乱问题的用户。","2026-06-11 03:46:07","high_star"]