ollama

Use it when

• run open source models fully offline on a laptop
• local LLM inference without cloud API keys
• airgapped model serving on my machine

What it solves

Not the fit when

• frontier model quality always

Install

curl -fsSL https://ollama.com/install.sh | sh  # or brew install ollama

Invoke

ollama pull llama3.2; ollama run llama3.2  # local inference to cut API cost and rate limits.

Alternatives

No reviewed alternatives recorded yet.