Mastering Ollama CLI Commands

Let us master the Ollama command-line interface (CLI) to pull, run, and manage local models on your computer.

1. Pulling Models Without Running

To download a model from the Ollama library registry to your disk without immediately opening a chat interface, use the pull command:

# Download the lightweight Llama 3 8B model
ollama pull llama3

2. Launching Interactive Chat Terminal

To run a model and open a console chat interface directly in your command line, run:

# Start and converse with Qwen2.5
ollama run qwen2.5

When inside the interactive session:

• Type your query and press Enter to get a response.
• Type /? to view available commands.
• Type /bye and press Enter to exit the session.

3. Catalog and Disk Cleanups

Listing Installed Models

To list all models currently downloaded on your system:

ollama list

Removing Models

To delete a model and free up space on your hard drive:

ollama rm llama3

Checking Model Architecture Details

To display configuration information, license agreements, or tensor details for a model:

ollama show qwen2.5