🦙 g023's OllamaMan - Ollama Manager

A comprehensive web-based dashboard for managing remote Ollama language model servers. This application provides an intuitive interface for model management, interactive chat, text generation, embeddings visualization, and server monitoring.

📋 Overview

g023's OllamaMan - Ollama Manager is designed to simplify the management and utilization of Ollama servers running on local networks or remote machines. It offers a unified web interface that eliminates the need for command-line interactions, making AI model deployment and usage accessible to users who prefer graphical interfaces.

What is Ollama?

Ollama is an open-source platform that enables running large language models locally on your hardware. It supports various model architectures and provides a REST API for programmatic access. Ollama Manager acts as a user-friendly frontend to this API, providing features like:

• Model Management: Browse, download, organize, and create custom AI models
• Interactive Chat: Conversational interface with advanced features
• Text Generation: Single-prompt completion tasks
• Embeddings: Vector representation generation and visualization
• Server Monitoring: Real-time status and performance metrics
• Model Customization: Create personalized models with fine-tuned parameters

✨ Key Features

Core Functionality

📊 Dashboard

• Server Status Monitoring: Real-time connection status and latency metrics
• Model Statistics: Overview of installed and running models
• Storage Information: Disk usage tracking for model storage
• Quick Actions: One-click access to common operations

📦 Model Manager & Creator

• Model Library: Browse available models from the connected Ollama server
• Download Management: Pull models with progress tracking
• Model Operations: Delete, copy, and inspect model details
• Search and Filter: Find models by name or capabilities
• Advanced Model Creation: Create custom models with modified parameters
• Parameter Tuning: Adjust temperature, context size, sampling settings, and more
• System Prompts: Define AI behavior and personality
• Prompt Templates: Customize prompt formatting
• Message Examples: Add conversation examples for fine-tuning
• Stop Sequences: Define custom stopping conditions
• Configuration Import/Export: Save and load model configurations
• Preview Mode: Review model settings before creation

🛠️ Model Creator

• Custom Model Building: Create personalized AI models from existing base models
• Parameter Modification: Fine-tune model behavior with comprehensive parameter controls:
  • Core Parameters: Temperature, context size, token limits
  • Sampling Controls: Top-k, top-p, min-p, typical-p for generation quality
  • Repetition Management: Penalty settings and sequence controls
• System Prompt Integration: Define AI personality and behavior patterns
• Template Customization: Control prompt formatting and structure
• Example Conversations: Add training examples for specialized behaviors
• Stop Sequence Management: Define custom termination conditions
• Parameter Comparison: Visual diff showing changes from source model
• Configuration Management: Import/export model configurations as JSON
• Preview & Validation: Review complete model setup before creation
• Advanced Workflows: Support for complex model configurations and fine-tuning

💬 Advanced Chat Interface

• Conversational AI: Full-featured chat with context preservation
• Streaming Responses: Real-time text generation for immediate feedback
• System Prompts: Customize AI behavior with predefined personas
• Parameter Tuning: Adjust temperature, token limits, and sampling settings
• Chat History: Persistent conversation storage and retrieval

✨ Text Generation

• Prompt Engineering: Single-turn text completion
• Model Selection: Choose from available models
• Parameter Control: Fine-tune generation settings
• Output Formatting: Clean display with syntax highlighting

🎯 Embeddings Playground

• Vector Generation: Convert text to numerical representations
• Visualization: Interactive charts showing embedding distributions
• Statistical Analysis: Mean, min/max values, and dimensionality info
• Model Compatibility: Works with embedding-specialized models

📋 API Logs

• Request Tracking: Monitor all API interactions
• Performance Metrics: Response times and error logging
• Debugging Tools: Inspect request/response data
• Log Management: Clear and filter log entries

Advanced Features

⚖️ Model Comparison

• Side-by-Side Analysis: Compare outputs from different models
• Performance Metrics: Token counts and generation times
• A/B Testing: Evaluate model responses for specific tasks

🖥️ Integrated Terminal

• Command Interface: Direct access to Ollama CLI commands
• Output Display: Formatted terminal output in the web interface
• Command History: Recall and reuse previous commands

⚙️ Comprehensive Settings

• Theme Selection: Light and dark interface modes
• Server Configuration: Dynamic Ollama host and port settings
• Default Preferences: Set default models and parameters
• Auto-Refresh: Configurable status update intervals
• Notification Controls: Manage alert preferences

🔍 Spotlight Search

• Global Navigation: Quick access to all features
• Model Search: Find models across the registry
• Command Shortcuts: Keyboard-driven navigation

🎨 User Interface

• Window Management: Draggable, resizable interface windows
• Notification System: Toast-style alerts for user feedback
• Responsive Design: Adapts to different screen sizes
• Accessibility: Keyboard navigation and screen reader support

🚀 Installation

System Requirements

• Web Server: Apache, Nginx, or any PHP-compatible server
• PHP Version: 7.4 or higher
• PHP Extensions: cURL (required for API communication)
• Ollama Server: Running instance accessible via HTTP
• Browser: Modern web browser with JavaScript enabled

Quick Setup

1. Download the Application

   # Clone or download to your web server's document root
   # Example paths:
   /var/www/html/ollama-manager/
   # or for local development:
   C:\xampp\htdocs\ollama-manager\
   

2. Configure Server Access

   Edit api/config.php to set your Ollama server details:

   define('OLLAMA_HOST', '192.168.1.100');  // Your Ollama server IP
   define('OLLAMA_PORT', '11434');         // Default Ollama port

3. Server Configuration

   Ensure your Ollama server accepts remote connections:

   # On the Ollama server machine
   OLLAMA_HOST=0.0.0.0 ollama serve

4. Access the Interface

   Open your web browser and navigate to:

   http://localhost/ollama-manager/
   

Directory Permissions

Ensure the web server can write to the data/ directory for storing settings and logs:

chmod 755 data/

Database Initialization

The application uses SQLite for data storage. The database is automatically created when first accessed. For manual initialization or migration from older JSON-based storage:

# Run the database initialization script
php api/init_db.php

This will create the database schema and migrate any existing JSON data to SQLite.

⚙️ Configuration

Server Connection

The application connects to Ollama via HTTP API. Key configuration options in api/config.php:

• OLLAMA_HOST: IP address or hostname of the Ollama server
• OLLAMA_PORT: Port number (default: 11434)
• OLLAMA_TIMEOUT_*: Various timeout settings for different operations

Application Settings

Accessible through the Settings window or api/settings.php:

• Server Settings:
  • Ollama Host: IP address or hostname of your Ollama server
  • Ollama Port: Port number (default: 11434)
• Theme: Interface appearance (light/dark)
• Default Model: Pre-selected model for chat and generation
• Auto-refresh Interval: Status update frequency
• Notification Preferences: Alert display settings

📖 Usage Guide

Getting Started

1. Verify Connection: Check the server status indicator in the top menu bar
2. Load Models: Use the Model Manager to browse and download models
3. Start Chatting: Open the Chat window and select a model
4. Explore Features: Try text generation and embeddings playground

Model Management

Downloading Models

• Navigate to Model Manager window
• Browse or search for models
• Click "Pull Model" to download
• Monitor progress in the interface

Model Information

• Select a model to view details
• See size, modification date, and capabilities
• Copy or delete models as needed

Creating Custom Models

• Open the Model Creator window
• Select a source model as the base
• Modify parameters using the intuitive interface:
  • Adjust core parameters (temperature, context size, etc.)
  • Configure sampling settings for generation quality
  • Set repetition controls to avoid loops
  • Add stop sequences for precise control
• Define system prompts for AI personality
• Customize prompt templates
• Add example conversations for fine-tuning
• Preview the complete configuration
• Create the model with one click
• Import/export configurations for reuse

Chat Interface

Basic Conversation

• Select a model from the dropdown
• Type messages in the input field
• Press Enter or click send button
• Responses appear in real-time with streaming

Advanced Features

• System Prompts: Set AI behavior with custom instructions
• Parameters: Adjust creativity and response length
• Image Input: Upload images for vision-capable models
• Tool Calling: Enable function calling for enhanced capabilities
• Structured Output: Generate JSON responses with schemas

Chat History

• Conversations are automatically saved
• Access previous chats from the sidebar
• Search through chat history
• Export conversations for external use

Text Generation

Single Prompts

• Choose a model and enter a prompt
• Adjust generation parameters
• Generate and view formatted output
• Useful for creative writing, code generation, etc.

Embeddings Visualization

Vector Analysis

• Select an embedding model
• Input text for vector conversion
• View statistical properties
• Visualize embedding distributions

Monitoring and Logs

Server Status

• Real-time connection monitoring
• Performance metrics display
• Running model tracking

API Logs

• View all API requests and responses
• Debug failed operations
• Monitor usage patterns

🔧 API Reference

The application provides a REST API for programmatic access. All endpoints return JSON responses.

Response Format

{
  "success": true|false,
  "data": { ... } | null,
  "error": "message" | null
}

Available Endpoints

Endpoint	Method	Description
/api/status.php	GET	Server status and statistics
/api/models.php?action=list	GET	List available models
/api/models.php?action=show&model=name	GET	Get model details
/api/models.php?action=pull	POST	Download a model
/api/models.php?action=delete	POST	Remove a model
/api/models.php?action=copy	POST	Duplicate a model
/api/models.php?action=create_advanced	POST	Create custom model with parameters
/api/chat.php	POST	Send chat message
/api/generate.php	POST	Generate text completion
/api/embed.php	POST	Create embeddings
/api/settings.php?action=get	GET	Retrieve settings
/api/settings.php?action=save	POST	Update settings
/api/history.php?action=list	GET	Get chat history
/api/logs.php?action=list	GET	View API logs

Example API Usage

Chat Request

fetch('api/chat.php', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    model: 'llama2',
    messages: [{ role: 'user', content: 'Hello!' }]
  })
})
.then(response => response.json())
.then(data => console.log(data));

🛠️ Development

Architecture

The application follows a client-server architecture:

• Frontend: HTML, CSS, JavaScript (jQuery)
• Backend: PHP API layer
• Database: SQLite for data persistence
• Styling: Custom CSS with theme support

Project Structure

ollama-manager/
├── index.php              # Main application interface
├── api/                   # Backend API endpoints
│   ├── config.php         # Configuration constants
│   ├── ollama.php         # Ollama API wrapper
│   ├── database.php       # Data persistence layer
│   ├── models.php         # Model management and creation API
│   └── *.php              # Individual API handlers
├── assets/                # Static resources
│   ├── css/               # Stylesheets (Aqua UI theme)
│   ├── js/                # JavaScript logic (Model Creator, Chat, etc.)
│   └── images/            # Interface graphics
├── data/                  # Application data storage
    └── uploads/           # File upload directory

Key Technologies

• PHP: Server-side logic and API implementation
• SQLite: Lightweight database for settings and history
• jQuery: DOM manipulation and AJAX communication
• Chart.js: Data visualization (embeddings)
• Highlight.js: Syntax highlighting for code
• Marked.js: Markdown rendering

Extending the Application

Adding New Features

1. Create API endpoint in api/ directory
2. Add frontend JavaScript in assets/js/app.js
3. Update HTML interface in index.php
4. Add database schema changes if needed

Custom Themes

• Modify CSS variables in assets/css/
• Create new theme files following existing patterns
• Update theme selection in settings

📄 License

This project is licensed under the BSD 3-Clause License - see the LICENSE file for details.

Copyright (c) 2025, g023

👤 Author

g023

• GitHub: g023
• X (Twitter): @g023dev

🙏 Acknowledgments

• Built for the Ollama community
• Inspired by the need for accessible AI model management

---

Enjoy managing your AI models with Ollama Manager! 🦙✨