Ollama: Run AI Locally for Free — No API Costs, No Data Leaks

What Is Ollama and Why Does It Matter?

Hardware Requirements

Step-by-Step Installation

macOS

# Option 1: Download the app (recommended for non-technical users)
# Visit https://ollama.com/download and download the macOS app
# Drag to Applications, launch it — Ollama runs as a menu bar app

# Option 2: Install via Homebrew
brew install ollama

# Start the Ollama service
ollama serve &

# Pull and run your first model
ollama run llama3.2

# Once the model downloads, you'll get an interactive prompt
# Type your message and press Enter

Windows

# Download the Windows installer from https://ollama.com/download
# Run the .exe installer — it adds Ollama to your system tray

# Open PowerShell or Command Prompt, then:
ollama run llama3.2

# Ollama runs as a background service automatically after install
# Access it at http://localhost:11434

Linux (Ubuntu/Debian)

# One-line install script (official)
curl -fsSL https://ollama.com/install.sh | sh

# Start the service
sudo systemctl start ollama
sudo systemctl enable ollama   # auto-start on boot

# Verify it's running
curl http://localhost:11434

# Pull your first model
ollama pull llama3.2
ollama pull mistral

Running Your First Model

# Interactive chat
ollama run llama3.2

# You'll see download progress, then a >>> prompt
>>> Write a 5-star Google review reply for a coffee shop

# To exit the chat
>>> /bye

# Pull additional models
ollama pull mistral
ollama pull phi4-mini
ollama pull gemma2:9b

# List downloaded models
ollama list

# Remove a model you no longer need (frees up disk space)
ollama rm llama3.2

Python Integration: Calling Ollama's Local API

import requests
import json

OLLAMA_URL = "http://localhost:11434/api/generate"
OLLAMA_CHAT_URL = "http://localhost:11434/api/chat"

def ollama_generate(prompt: str, model: str = "llama3.2") -> str:
    """Simple single-turn generation."""
    response = requests.post(
        OLLAMA_URL,
        json={
            "model": model,
            "prompt": prompt,
            "stream": False,   # Set True for streaming output
        }
    )
    response.raise_for_status()
    return response.json()["response"]

def ollama_chat(messages: list, model: str = "mistral") -> str:
    """Multi-turn chat with message history."""
    response = requests.post(
        OLLAMA_CHAT_URL,
        json={
            "model": model,
            "messages": messages,
            "stream": False,
        }
    )
    response.raise_for_status()
    return response.json()["message"]["content"]

# Using the OpenAI SDK with Ollama (drop-in replacement)
from openai import OpenAI

client = OpenAI(
    base_url="http://localhost:11434/v1",
    api_key="ollama",   # Required by SDK but not used by Ollama
)

response = client.chat.completions.create(
    model="llama3.2",
    messages=[{"role": "user", "content": "Write a Monday promotion caption for a yoga studio"}]
)
print(response.choices[0].message.content)

4 Practical Automations for Local Businesses

Automation 1: Offline Review Reply Generator

import csv
import requests
import json
import time

OLLAMA_URL = "http://localhost:11434/api/generate"
BUSINESS_NAME = "The Brow Studio"
BUSINESS_TYPE = "eyebrow and lash salon"

def generate_review_reply(review_text: str, rating: int) -> str:
    tone = "warm and grateful" if rating >= 4 else "apologetic and solution-focused"
    prompt = f"""You are writing a Google review reply for {BUSINESS_NAME}, a {BUSINESS_TYPE}.
Write a {tone} reply (2-3 sentences) to this {rating}-star review.
Keep it professional and human. Don't start with "Thank you for your review."

Review: {review_text}

Reply:"""

    response = requests.post(
        OLLAMA_URL,
        json={"model": "mistral", "prompt": prompt, "stream": False}
    )
    return response.json()["response"].strip()

def process_reviews_csv(input_file: str, output_file: str):
    results = []
    with open(input_file, "r") as f:
        reader = csv.DictReader(f)
        for i, row in enumerate(reader):
            print(f"Processing review {i+1}...")
            reply = generate_review_reply(row["review_text"], int(row["rating"]))
            results.append({**row, "suggested_reply": reply})
            time.sleep(0.5)  # Small pause between calls

    with open(output_file, "w", newline="") as f:
        writer = csv.DictWriter(f, fieldnames=results[0].keys())
        writer.writeheader()
        writer.writerows(results)
    print(f"Done! Replies saved to {output_file}")

# Usage:
# process_reviews_csv("reviews.csv", "reviews_with_replies.csv")
# Input CSV needs columns: review_text, rating

Automation 2: Local FAQ Chatbot Using a Text File as Knowledge Base

import requests

def load_knowledge_base(filepath: str) -> str:
    with open(filepath, "r") as f:
        return f.read()

def answer_customer_question(question: str, knowledge_base: str, model: str = "llama3.2") -> str:
    prompt = f"""You are a helpful assistant for a local business. Answer the customer's question
using ONLY the information in the knowledge base below. If the answer isn't in the knowledge base,
say "I'm not sure — please call us or check our website."

KNOWLEDGE BASE:
{knowledge_base}

CUSTOMER QUESTION: {question}

ANSWER:"""

    response = requests.post(
        "http://localhost:11434/api/generate",
        json={"model": model, "prompt": prompt, "stream": False}
    )
    return response.json()["response"].strip()

# Create a file called business_faq.txt with content like:
# Hours: Monday-Saturday 9am-7pm, Sunday 10am-5pm
# Address: 123 Main Street, Austin TX 78701
# Parking: Free street parking on Oak Ave
# Services: Haircuts ($45), Color ($90+), Blowouts ($35)
# Booking: Call (512) 555-0123 or book online at example.com

kb = load_knowledge_base("business_faq.txt")
answer = answer_customer_question("What time do you open on Sundays?", kb)
print(answer)

Automation 3: Batch Social Media Caption Generator

import csv
import requests

def generate_caption(service_name: str, promo_detail: str, platform: str = "Instagram") -> str:
    prompt = f"""Write a {platform} caption for a local salon promoting: {service_name}.
Promotion detail: {promo_detail}
Include 4-5 relevant hashtags. Keep it under 150 words. Be warm and engaging."""

    response = requests.post(
        "http://localhost:11434/api/generate",
        json={"model": "llama3.2", "prompt": prompt, "stream": False}
    )
    return response.json()["response"].strip()

def batch_generate_captions(input_csv: str, output_csv: str):
    with open(input_csv, "r") as f:
        rows = list(csv.DictReader(f))

    for row in rows:
        row["caption"] = generate_caption(row["service"], row["promo"])
        print(f"Generated caption for: {row['service']}")

    with open(output_csv, "w", newline="") as f:
        writer = csv.DictWriter(f, fieldnames=rows[0].keys())
        writer.writeheader()
        writer.writerows(rows)

# Input CSV example:
# service,promo
# "Balayage","20% off this weekend only"
# "Keratin Treatment","Book before June 30 for $20 off"
# "Lash Lift + Tint","New service — introductory price $65"

Automation 4: Email Draft Writer with Your Business's Tone of Voice

import requests

TONE_DESCRIPTION = """
Warm and professional. We run a family-owned pet grooming salon.
We're friendly and personal — we know every dog by name.
We're not corporate. We use simple language and genuine warmth.
We occasionally use light humor but never sarcasm.
"""

def draft_business_email(
    email_type: str,
    recipient_context: str,
    key_points: list[str]
) -> str:
    points_formatted = "\n".join(f"- {p}" for p in key_points)
    prompt = f"""Draft a business email for a pet grooming salon.

Our brand voice: {TONE_DESCRIPTION}

Email type: {email_type}
Recipient context: {recipient_context}
Key points to include:
{points_formatted}

Write the full email including subject line. Keep it under 200 words."""

    response = requests.post(
        "http://localhost:11434/api/generate",
        json={"model": "mistral", "prompt": prompt, "stream": False}
    )
    return response.json()["response"].strip()

# Example usage
email = draft_business_email(
    email_type="Appointment reminder",
    recipient_context="Regular customer, Golden Retriever named Biscuit, monthly groom",
    key_points=[
        "Appointment tomorrow at 2pm",
        "Reminder to bring vaccination records if not already on file",
        "New parking: use the side entrance off Maple St"
    ]
)
print(email)

Limitations You Should Know

FAQ

Related Articles

• Llama 4 for Local Business Automation — Free on Groq
• OpenRouter Complete Guide for Local Business Automation 2026
• GPT-4o Mini vs Groq Llama: Real Cost Comparison for Business AI
• Best Free Models on OpenRouter for Business Automation 2026
• DeepSeek R2 vs Claude for Local Business — Which Wins in 2026?