A command-line interface for chatting with AI models via OpenRouter, supporting model selection, streaming responses, file attachments, credit tracking, and configurable settings.

Features

Select and switch between OpenRouter models
Streaming or buffered responses
Attach images and text files to messages
View credit usage and balance
Set persistent default model
Keyboard shortcuts and history

Installation

Requires Python 3.8+. Download the repository. Install dependencies:

pip install typer rich openrouter pyperclip requests prompt_toolkit

Or use :

pip install -r requiremnets.txt

For convience you can move the application to a folder in you $PATH and rename it to just oai .

Usage

Run the app and interact via commands:

python oai_chat.py

Example session:

Welcome to oAI!
You> /model gpt
(Shows table of GPT models)
Enter model number: 1
Selected: GPT-4o (openai/gpt-4o)

You> Hello, how are you?
Streaming response... (Press Ctrl+C to cancel)
Hello! I'm doing great...

For full commands, use /help in the app.

Screenshot

License

This project is licensed under the MIT License - see the MIT License.

Author: Rune Olsen