ModelMeter

ModelMeter is a lightweight, cross-platform utility that calculates the total storage footprint of locally installed large language models managed through Ollama.

📋 Overview

Managing multiple large language models locally can quickly consume significant disk space. ModelMeter provides a simple command-line tool to instantly determine exactly how much storage your LLM collection is using, helping you manage system resources effectively.

🚀 Quick Start

# Python
./py/ollama_total.py

# Deno (JavaScript)
deno run --allow-run ./js/ollama_total.js

# Go
./go/ollama_total

# Example output
26.8 GB

✨ Features

• Accurate Size Calculation: Precisely measures total storage used by all installed Ollama models
• Smart Unit Formatting: Automatically selects the most appropriate unit (MB, GB, TB)
• Multiple Language Implementations: Choose from Python, Deno (JavaScript), or Go
• Simple Integration: Use as a standalone command or integrate into larger workflows
• Minimal Dependencies: Relies only on standard libraries in each language

❓ Why ModelMeter?

While you could manually add up model sizes from ollama list output, ModelMeter:

• Saves time with instant calculations
• Eliminates human error in size conversions and addition
• Provides consistent formatting across different model sizes
• Enables easy integration into monitoring scripts and dashboards
• Serves as a building block for more complex LLM resource management tools

💻 Installation

System Requirements

• Python: Python 3.10+
• JavaScript: Deno 2.0+
• Go: Go 1.16+
• All implementations: Ollama must be installed and in your PATH

Python Implementation

# Clone the repository
git clone https://github.com/ai-mindset/ModelMeter.git
cd ModelMeter/py

# Make the Python script executable
chmod +x ollama_total.py

# Optional: Add to your PATH
ln -s $(pwd)/ollama_total.py /usr/local/bin/ollama_total

Deno Implementation

# Clone the repository
git clone https://github.com/ai-mindset/ModelMeter.git
cd ModelMeter/js

# Make the JavaScript file executable
chmod +x ollama_total.js

# Run with Deno
deno run --allow-run ollama_total.js

# Optional: Install as a command
deno install --allow-run --name ollama_total ollama_total.js

Go Implementation

# Clone the repository
git clone https://github.com/ai-mindset/ModelMeter.git
cd ModelMeter/go

# Build the Go binary
go build -o ollama_total ollama_total.go

# Move to a directory in your PATH
sudo mv ollama_total /usr/local/bin/

💻 Usage

Basic Command

ollama_total

Integrating Into Scripts

Python

from ollama_total import get_ollama_total

# Get the total storage used by Ollama models
total = get_ollama_total()
print(f"Your Ollama models are using {total}")

Deno (JavaScript)

import { getOllamaTotal } from "./ollama_total.js";

// Get the total storage used by Ollama models
const total = await getOllamaTotal();
console.log(`Your Ollama models are using ${total}`);

Go

package main

import (
    "fmt"
    "path/to/modelMeter"
)

func main() {
    total, err := getOllamaTotal()
    if err != nil {
        fmt.Printf("Error: %s\n", err)
        return
    }
    fmt.Printf("Your Ollama models are using %s\n", total)
}

🔍 Technical Details

How It Works

ModelMeter executes the ollama list command, parses the output to extract size information, converts all sizes to bytes for accurate summation, and then converts the total back to the most appropriate unit.

Size Conversion and Parsing

• Uses SI units: 1 MB = 10^6 bytes, 1 GB = 10^9 bytes, 1 TB = 10^12 bytes (matching Ollama's output)
• Regex pattern (\d+(?:\.\d+)?) (MB|GB|TB) extracts size values and units
• Results are formatted with one decimal place precision for readability

⚠️ Troubleshooting

Issue	Solution
Command not found: ollama	Ensure Ollama is installed and in your PATH
No output displayed	Check if you have any models installed with ollama list
Permission denied	Run chmod +x on the script file
Incorrect total	Ensure no other process is installing/removing models during calculation
Deno security errors	Ensure you've provided the --allow-run flag when using Deno

🔄 Alternatives

Approach	Pros	Cons
ModelMeter	Automatic calculation, multiple languages, formatting	Additional tool to install
Manual addition from ollama list	No installation needed	Time-consuming, error-prone
Shell one-liner1	Quick for one-time use	Complex syntax, limited formatting
Disk usage tools (du)	System-wide analysis	Includes non-model files, less precise

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Ollama - For making local LLM deployment accessible

---

Footnotes

1. Example shell one-liner: ollama list | awk 'NR>1 {size=$3; if($4=="GB") size*=1000; if($4=="TB") size*=1000000; sum+=size} END {printf "%.1f %s\n", sum>=1000000 ? sum/1000000 : sum>=1000 ? sum/1000 : sum, sum>=1000000 ? "TB" : sum>=1000 ? "GB" : "MB"}' ↩