How to beginner · 3 min read

How to run local AI assistant on PC

Q: How to run local AI assistant on PC

Use Ollama to run AI models locally on your PC by installing the Ollama CLI and downloading supported models. You can then start a local AI assistant by running ollama run with your chosen model, enabling private and offline AI interactions.

Quick answer

Use Ollama to run AI models locally on your PC by installing the Ollama CLI and downloading supported models. You can then start a local AI assistant by running ollama run with your chosen model, enabling private and offline AI interactions.

PREREQUISITES

Windows 10/11 or macOS
Python 3.8+ (optional for scripting)
Ollama CLI installed
Internet connection for initial model download

Setup Ollama CLI

Install the Ollama CLI to manage and run AI models locally on your PC. Ollama supports Windows and macOS. Download the installer from the official Ollama website and follow the installation instructions.

After installation, verify by running ollama version in your terminal or command prompt.

bash

ollama version

output

Ollama CLI version 1.0.0

Run a local AI assistant

Once Ollama CLI is installed, you can run a local AI assistant by pulling a model and starting an interactive session.

Example: Run the llama2 model locally with:

bash

ollama run llama2

output

Ollama Llama2 model loaded.
You: Hello, AI assistant!
AI: Hello! How can I help you today?

Step by step Python integration

You can also integrate Ollama models into Python scripts using subprocess calls to run the local assistant and capture responses.

python

import subprocess

# Run Ollama local model and send a prompt
command = ['ollama', 'run', 'llama2', '--prompt', 'What is the weather today?']
result = subprocess.run(command, capture_output=True, text=True)
print(result.stdout)

output

AI: The weather today is sunny with a high of 75°F.

Common variations

Use different models like llama2, mistral, or falcon by specifying the model name in ollama run.
Run Ollama in server mode for API access with ollama serve.
Integrate with Python asynchronously using asyncio.create_subprocess_exec for non-blocking calls.

Troubleshooting

If ollama run fails, ensure the model is downloaded by running ollama pull <model-name>.
Check your PATH environment variable includes the Ollama CLI directory.
For permission errors, run the terminal as administrator or with elevated privileges.

✅

Key Takeaways

Install Ollama CLI to run AI models locally on your PC without internet after initial setup.
Use ollama run <model> to start an interactive local AI assistant session.
Integrate Ollama with Python via subprocess calls for custom local AI workflows.

Verified 2026-04 · llama2, mistral, falcon

Verify ↗