Explore running models on your computer or through supported services, checking which mode the task actually uses.
Where it fits
Ollama’s documentation includes local models, cloud models, integrations, and APIs. “Using Ollama” is therefore not enough to establish where a particular run happens; confirm the selected model and mode.
A sensible first task
Try an appropriate local model with non-sensitive sample text. Compare output quality and response time on a small task, then inspect any server or integration you enable.
What to check
Review the actual account, feature, and output for the task you intend to do.
- Local versus cloud execution for the selected model.
- Model license, publisher, hardware needs, and download source.
- Network-facing endpoints, logs, extensions, and updates.
How it connects to the bigger picture
It is a runtime layer for model use. Local computation can change the information path, but privacy and security still depend on the entire configuration.
Sources and verification
Ollama documentation · Ollama