Agentic LLM

Free, local AI on your own hardware.

Ollama runs Llama 3, Mistral, Gemma, and TinyLlama on your GPU or CPU. Zero API bills, fully offline, unlimited queries. Requires the home server — the cloud-only operator hits Claude or Kimi instead.

Choose Your Plan Back to overview

Free local AI via Ollama — Llama 3, Mistral, Gemma, TinyLlama on your own GPU or CPU.

What it does

Ollama LLM at a glance.

Always-tried-first

When the home server runs Ollama, OB1 routes free queries through it before falling back to a paid cloud brain. Energy cost stays at zero.

Pick your model

Llama 3 8B for general chat, Mistral 7B for code, Gemma 2 9B for reasoning, TinyLlama 1.1B for low-end laptops. Mix per task.

GPU or CPU

Modern Apple Silicon and consumer NVIDIA cards run 7B models comfortably. CPU-only fallback works for lighter models on older boxes.

Auto-pulled

Pick a model in the Ollama LLM app; the installer downloads weights and warms them. Zero command-line.

Add Ollama LLM to your agent.

Every VP3 plan unlocks the full c:// drive catalog. Pick a plan, install the home server, flip the app on.

Choose Your Plan