OmniChat CLI 🤖

A powerful CLI tool that combines multiple AI models (OpenAI, Groq) for chat, image generation, and content creation. Features streaming responses, conversation history, and multi-format exports.

📚 View Full Documentation

✨ Features

AI Providers

• 🧠 OpenAI (gpt-3.5-turbo)
• 🚀 Groq (llama3-8b-8192)
• 🎨 Image Generation

Core Features

• 📡 Real-time streaming responses
• 🌈 Colorful, emoji-enhanced interface
• 💾 Chat history saving with custom filenames
• 📁 Custom save locations
• 📊 Multiple export formats
  • JSON (structured data)
  • PDF (formatted document)
  • Markdown (human-readable)
• ⚙️ Configurable parameters
  • Temperature control
  • Token limits
  • Model selection
  • Response streaming
  • Export format selection

Interface

• 👤 User messages in yellow
• 🤖 Assistant responses in blue
• 🎨 Image generation with custom prompts
• 🚦 Color-coded status messages
• 📝 Easy-to-read format with emoji indicators

🛠️ Prerequisites

• Python 3.10+
• Required API Keys:
  • OpenAI API key
  • Groq API key
  • Image Generation API URL

📦 Installation

From PyPI

pip install omenicli

From Source

git clone https://github.com/Amul-Thantharate/omenicli.git
cd omenicli
pip install -e .

3. Configure API Keys:

Method A: Environment Variables

# Windows (Command Prompt)
set OPENAI_API_KEY=your_openai_api_key
set GROQ_API_KEY=your_groq_api_key
set APP_URL=your_image_generation_api_url

# Windows (PowerShell)
$env:OPENAI_API_KEY="your_openai_api_key"
$env:GROQ_API_KEY="your_groq_api_key"
$env:APP_URL="your_image_generation_api_url"

# Linux/Mac
export OPENAI_API_KEY="your_openai_api_key"
export GROQ_API_KEY="your_groq_api_key"
export APP_URL="your_image_generation_api_url"

Method B: .env File

Create a .env file in the project root:

OPENAI_API_KEY=your_openai_api_key
GROQ_API_KEY=your_groq_api_key
APP_URL=your_image_generation_api_url

🚀 Usage

Quick Start

1. Chat with OpenAI (Default):

omenicli

2. Chat with Groq:

omenicli --model-type groq

3. Generate Images:

omenicli --model-type image

Advanced Usage

1. Stream Responses:

omenicli --stream

2. Custom Model Settings:

omenicli --temperature 0.7 --max-tokens 2048

3. Save Chat History:

omenicli --save

4. Custom Image Directory:

omenicli --model-type image --image-dir my_images

5. Combined Features:

omenicli --model-type image --save --image-dir my_images --stream

📚 Example

Refer to the Example section.

⚙️ Configuration Options

Option	Short	Default	Description
--model-type	-mt	openai	Model provider (openai/groq/image)
--temperature	-T	0.5	Response randomness (0-1)
--max-tokens	-M	1024	Maximum response length
--stream	-S	False	Enable streaming responses
--save	-s	False	Save chat history
--openai-model	-o	gpt-3.5-turbo	OpenAI model name
--groq-model	-g	llama3-8b-8192	Groq model name
--image-dir	-i	generated_images	Directory to save generated images
--export-format	-e	json	Export format (json/pdf/markdown)

💾 Data Management

Chat History

• Default Location: chat_history/
• Format: JSON files
• Custom Options:
  • Directory: Enter custom path when prompted
  • Filename: Enter custom name when prompted
  • Default naming: chat_[model-type]_[timestamp].json

Generated Images

• Default Location: generated_images/
• Format: PNG files with timestamps
• Custom Locations:
  • Via CLI: --image-dir path/to/directory
  • Via Prompt: Enter path when asked
• Naming: image_[timestamp].png

🔧 Error Handling

The application handles various scenarios:

• Missing/Invalid API keys
• Network connectivity issues
• Rate limiting
• Invalid configurations
• File system errors

🤝 Contributing

Feel free to:

• Open issues
• Submit pull requests
• Suggest improvements
• Report bugs

📝 License

MIT License

🔍 Tips & Tricks

1. Chat Mode:

   • Use "exit" to end the session
   • Try different temperatures for varied responses
   • Enable streaming for real-time responses
   • Use custom filenames for better organization

2. Image Generation:

   • Be specific in your prompts
   • Use custom directories for organization
   • Combine with chat history saving

3. Chat History:

   • Use descriptive filenames
   • Organize by project/topic
   • Use custom paths for better file management