Skip to Content

Ollama setup

Ollama pull downloads a model; ollama run starts an interactive session; ollama serve exposes the OpenAI-compatible API on port 11434.

Hub: Ollama.

Pull and run

ollama pull llama3.2 ollama run llama3.2 # Loaded models and memory ollama ps # API server (default :11434) ollama serve

Model sizing by VRAM

VRAMPractical modelsFit
8 GB3B–7B Q4Exploration
16 GB7B–13BMost IDE tasks
24 GB+14B–32B, 70B QHeavier work; still below frontier

Benchmark quants (Q4, Q5, Q8) on your prompts — Q4 that fails review costs more than API tokens saved.

Unload models

ollama stop llama3.2

Loading five models simultaneously causes VRAM thrashing and unpredictable latency.

Last updated on