LLM CLI

A simple command-line tool to interact with AI models using OpenRouter's API. It provides a minimal chat interface, stream-based responses, and basic conversation history management.

Key Features

Streamed real-time AI responses
Basic conversation management
Simple API key handling

Setup

Clone the repo and install dependencies:

git clone https://github.com/lunarzack/llm-cli.git

pip install requests python-dotenv

Set up your OpenRouter API key in a `.env` file or hardcode it.

Usage

Run the script with:

python llm-cli.py

Type your message, and the AI responds in real-time. Type exit to quit.

Configuration

Change the model in the script (default: "deepseek/deepseek-chat:free")
Adjust settings like temperature and max_tokens for different outputs.

Code Highlights

Handles secure API key management via environment variables
Manages streaming responses and chat loop
Parses JSON response chunks for proper output

Acknowledgements

Thanks to OpenRouter for the API and DeepSeek AI for the language models.