Fred Terzi
Fred Terzi's Totem LLM: a privacy-first AI agent that runs locally on consumer hardware, with Ollama/OpenRouter model selection and in-browser RAG.
Last verified:
What is Fred Terzi?
Fred Terzi maintains Totem LLM, a sovereign, privacy-first AI agent designed to run locally on consumer-grade hardware (laptops and desktops) without requiring cloud-based processing. Users keep all their data private on their own machine, with full control over AI interactions and no reliance on third-party services. The tool serves individuals and small teams who need data-sensitive AI applications, developers building private AI workflows, and anyone wanting to avoid vendor lock-in or cloud API costs.
Key features include privacy-first architecture where all data stays on the user's machine, optimization for consumer hardware without expensive GPU servers, and flexible model selection between local models via Ollama or cloud providers through OpenRouter. The system offers one-line installation with automatic dependency resolution, modular architecture for easy integration of new features, document ingestion and collection capabilities for RAG (Retrieval-Augmented Generation), and a browser-based UI accessible at localhost:8686. Users can also install it as a desktop app with its own window and taskbar icon.
Totem LLM is built on the foundation of AnythingLLM by Mintplex Labs and is released under the MIT License. It requires Node.js v18 or later and either Ollama for local models or an OpenRouter API key for cloud models. The tool supports Linux, macOS, and Windows platforms, with custom port configuration and customizable storage directory options available through environment variables.
Fred Terzi pricing
Pricing model: Freemium
Totem LLM itself is free and open source under the MIT License with no paid plans from the developer. For local models via Ollama, there is no additional cost beyond the free Ollama software. For cloud models through OpenRouter, users pay pass-through pricing at the same per-token rate as calling providers directly with a 5.5% credit purchase fee (minimum $0.80). OpenRouter offers 29 free models with no token cost but rate-limited to 20 requests/min and 50 requests/day (1,000/day after purchasing $10+ in credits). BYOK (Bring Your Own Key) mode offers first 1M requests/month free with 5% fee on provider cost beyond that.
Fred Terzi pros
- All data stays on your machine with complete privacy
- Optimized for consumer-grade laptops and desktops
- No expensive GPU servers required
- One-line npm install with automatic dependency resolution
- Choose between local Ollama models or cloud OpenRouter providers
- Modular architecture for easy feature integration
- Document ingestion for RAG capabilities
- Browser UI accessible at localhost:8686
- Can install as desktop app with own window
- Custom port configuration available
- Customizable storage directory via environment variable
- Works on Linux, macOS, and Windows
- No cloud-based processing required for local mode
- No vendor lock-in or third-party service reliance
- MIT License open source
Fred Terzi cons
- Requires Node.js v18 or later prerequisite
- Needs Ollama installed for local models or OpenRouter API key for cloud
- First run creates ~/totem-llm/ directory which may not suit all users
- Windows PowerShell users may need execution policy changes
- Windows users may need to use cmd.exe instead of PowerShell
- No official desktop app binary (browser-based link only)
- Document ingestion collector runs by default and may be unwanted
- No built-in mobile interface
- Cloud models still require paying OpenRouter credits
- Limited to single-machine deployment not multi-user server
Frequently asked questions about Fred Terzi
What is Totem LLM?
Totem LLM is a sovereign AI solution designed to run locally on consumer hardware while keeping your data private. It is a private AI agent that runs on your own hardware with your private data leading, offering full control over AI interactions without relying on third-party cloud services.
How do I install Totem LLM?
Open your terminal and run 'npm install -g totem-llm' which automatically downloads Totem LLM and installs all dependencies including server, collector, and frontend. Then run 'totem-llm start' to launch the service.
What hardware do I need to run Totem LLM?
Totem LLM is optimized for consumer-grade hardware like laptops and desktops. No expensive GPU servers are required. The minimum practical setup for local models is a GPU with 8GB VRAM running a small quantized model, though no GPU is strictly required as it can run on x86 CPU machines.
Can I use cloud models with Totem LLM?
Yes, while local AI is the core focus, Totem LLM supports both local and cloud-based LLM providers. You can use local models via Ollama or connect to cloud providers through OpenRouter using an API key, giving flexibility based on your needs and preferences.
How does Totem LLM handle my data privacy?
All data stays on your machine. Totem LLM is privacy-first by design, with 100% private operation where no data leaves your execution environment at any point. You have full control over your data and AI interactions without relying on third-party services.
What is the browser UI and how do I access it?
After starting Totem LLM with 'totem-llm start', open http://localhost:8686 in your browser to access the web GUI. The first run creates the ~/totem-llm/ directory, migrates the database, and launches all services automatically.
Can I install Totem LLM as a desktop application?
Yes, in the browser UI at localhost:8686, click the install icon in the address bar (or use your browser menu) to add Totem LLM to your desktop. This creates a link to the web GUI with its own window and taskbar/dock icon for easy access.
What are the system requirements for Totem LLM?
You need Node.js v18 or later installed. For local models, install Ollama (recommended). For cloud models, you need an OpenRouter API key. The tool runs on Linux, macOS, and Windows.
How do I customize the storage location for Totem LLM?
Override the default storage location (~/totem-llm) by setting the environment variable: export TOTEM_STORAGE_DIR=/your/custom/path before running 'totem-llm start'. This changes where all data and models are stored.
What command options are available for Totem LLM?
Available commands include: 'totem-llm start' to start server and collector, 'totem-llm --no-collector' to start server only without document ingestion, 'totem-llm --port=8080' for custom port, 'totem-llm setup' for first-time setup without starting, and 'totem-llm --help' to show all commands.