MOMOS (Mobile Models Ollama Setup) - Modern AI for Termux 🚀

A comprehensive installer for running AI models locally on Termux (Android) using Debian, Ollama, and various AI models with an interactive setup process.

📋 Table of Contents

• Features
• Requirements
• Installation Guide
• Usage Instructions
• Troubleshooting

✨ Features

• Modern Terminal UI with colors, emojis, and progress indicators
• Interactive Installation with step-by-step progress tracking
• Automatic Setup of Debian 12 via PRoot-Distro
• Ollama Integration with automatic installation and configuration
• Model Selection Menu with popular AI models
• TMUX Integration for background Ollama server management
• Error Handling with comprehensive error trapping
• Cross-Platform Compatibility for Termux on Android

🤖 Supported Models

The installer provides a selection menu with these options:

Pre-configured Models

• DeepSeek R1 1.5B - Fast, lightweight (recommended for mobile)
• DeepSeek R1 7B - Balanced performance and quality
• DeepSeek R1 14B - Higher quality, more resource intensive
• DeepSeek R1 32B - High quality, requires significant resources
• Custom Models - Any model from the Ollama library

Model Recommendations by Device Capability

• Low-end devices (2-3GB RAM): 1.5B models
• Mid-range devices (4-6GB RAM): 7B models
• High-end devices (8GB+ RAM): 14B+ models

📱 Requirements

Minimum Requirements

• Android 7.0+ (API level 24+)
• Termux (latest version from F-Droid or GitHub)
• 2GB free storage for basic models
• 3GB RAM minimum

Recommended Requirements

• Android 10+ for better performance
• 4GB+ RAM for larger models
• 5GB+ free storage for multiple models

Storage Requirements by Model

• 1.5B models: ~800MB
• 7B models: ~4GB
• 14B models: ~8GB
• 32B models: ~20GB

🚀 Installation Guide

Step 1: Install Termux

⚠️ IMPORTANT: Do NOT install Termux from Google Play Store

1. Download from F-Droid (Recommended):

   • Install F-Droid first, then search for "Termux"
   • Or download directly: https://f-droid.org/packages/com.termux/

2. Download from GitHub:

   • Visit: https://github.com/termux/termux-app/releases
   • Download the latest APK for your architecture

3. Install the APK and grant necessary permissions

Step 2: Setup Termux

# Update packages
pkg update && pkg upgrade -y

# Install essential tools
pkg install git wget curl -y

# Setup storage access
termux-setup-storage

Step 3: Clone Repository

# Clone the repository
git clone https://github.com/Sidharth-e/MOMOS.git
cd MOMOS

# Choose your setup script:
# For simple setup: chmod +x scripts/momo_setup.sh
# For advanced setup: chmod +x scripts/setup.sh

Step 4: Run the Installer

Option 1: Simple Setup (momo_setup.sh)

# Run the simple setup script
bash scripts/momo_setup.sh

Option 2: Advanced Setup (setup.sh)

# Run the advanced setup script with progress bars
bash scripts/setup.sh

What the Installer Does:

1. Updates Termux packages to latest versions
2. Installs PRoot-Distro for Linux distribution support
3. Installs Debian 12 with progress tracking
4. Presents Model Selection menu with options
5. Configures Debian environment
6. Installs Ollama automatically
7. Downloads Selected Model based on your choice
8. Sets up TMUX Session for background Ollama server
9. Provides Usage Instructions for next steps

Expected Time: 15-45 minutes depending on internet speed and model size

🎯 Usage Instructions

After Installation

Once the installer completes, you can use your AI models:

# Enter the Debian environment
proot-distro login debian

# Start chatting with your selected model
ollama run deepseek-r1:1.5b    # or whatever model you chose

TMUX Session Management

The installer automatically creates a TMUX session for the Ollama server:

# Enter Debian environment
proot-distro login debian

# Check if TMUX session exists
tmux list-sessions

# Attach to the ollama_server session
tmux attach-session -t ollama_server

# To detach (leave server running): Press CTRL+B then D

Managing Your Models

# Enter Debian environment
proot-distro login debian

# List installed models
ollama list

# Download additional models
ollama pull llama3:8b
ollama pull mistral:7b

# Remove models to free space
ollama rm model_name

# Get model information
ollama show model_name

🔧 Manual Commands

If you prefer to run commands directly:

# Enter Debian environment
proot-distro login debian

# Start Ollama server manually
ollama serve

# In another terminal, run any model
proot-distro login debian
ollama run deepseek-r1:1.5b    # Fast model
ollama run deepseek-r1:7b       # Balanced model
ollama run deepseek-r1:14b      # Higher quality model

🛠️ Troubleshooting

Common Installation Issues

1. "Permission denied" errors

# Solution: Make scripts executable
chmod +x scripts/*.sh

# Check permissions
ls -la scripts/

2. "Package not found" errors

# Solution: Update Termux packages
pkg update && pkg upgrade -y

# Refresh package lists
pkg update

3. "Storage access denied"

# Solution: Setup storage permissions
termux-setup-storage

# Grant permissions in Android Settings > Apps > Termux > Permissions

4. "Out of storage space"

# Check available space
df -h

# Clear Termux cache
pkg clean

# Remove unused packages
pkg autoremove

Installation-Specific Issues

1. "proot-distro not found" error

# Check if proot-distro is installed
which proot-distro

# If not found, install it
pkg install proot-distro -y

# Verify installation
proot-distro --help

2. "Debian installation failed" error

# Check available distributions
proot-distro list

# Try installing Debian again
proot-distro install debian

3. "Ollama installation failed" error

# Enter Debian environment
proot-distro login debian

# Install Ollama manually
curl -fsSL https://ollama.ai/install.sh | sh

# Verify installation
ollama --version

4. "Model download failed" error

# Enter Debian environment
proot-distro login debian

# Check internet connection
ping -c 3 google.com

# Try downloading the model again
ollama pull model_name

# Check available disk space
df -h

Performance Issues

1. Slow model loading

# Check available RAM
free -h

# Use smaller models for mobile devices
# Recommended: 1.5B, 7B, or 8B models

2. High memory usage

# Monitor memory usage
htop

# Kill unnecessary processes
pkill -f ollama

🔄 Updates

Updating Termux

pkg update && pkg upgrade -y

Updating Debian

proot-distro login debian -- bash -c "apt update && apt upgrade -y"

Updating Ollama

proot-distro login debian -- bash -c "curl -fsSL https://ollama.ai/install.sh | sh"

Updating Models

# Use the launcher's model selection
# Or manually:
proot-distro login debian -- bash -c "ollama pull model_name"

📞 Support

If you're still having issues:

1. Check the troubleshooting section above
2. Verify your Android version and Termux compatibility
3. Check Termux logs: logcat | grep termux
4. Ensure sufficient storage and RAM
5. Try a fresh Termux installation if all else fails
6. Check your internet connection during installation

---

Enjoy your modern AI experience on Termux! 🧠✨🤖

If this README helped you, consider giving it a star! ⭐