🧠 A minimalistic CLI agent for local LLM server

A lightweight and efficient command-line interface (CLI) agent designed to interact with any OpenAI-compatible API local server.

✨ Features

• ⚡ Fast CLI interaction with OpenAI-compatible APIs
• ⚡ Works with any local server implementing the OpenAI API spec (e.g. llama.cpp, vLLM, etc.)
• ⚡ Color console
• ⚡ Code syntax highlighting

🚀 Usage examples

Code highlighting

Table rendering

📦 Installation

git clone https://github.com/hardglitch/llm-client.git
cd llm-client
cargo build --release

⚙️ Configuration

-p, --port - server port (8080 by default)
-l, --log-file - log file ("log.log" by default)
--log-size - the log file size in bytes (104857600 by default)
--show-stat - show token stat (e.g. "[tokens: 97/ total: 543/ context: 16384]"), disabled by default

exit - exit from programm

📜 License

MIT