zeta-mlx-cli 0.3.2

CLI for Zeta MLX - OpenAI-compatible LLM/Embedding on Apple Silicon

zeta-mlx-cli

CLI for Zeta MLX - OpenAI-compatible LLM/Embedding inference on Apple Silicon.

Installation

pip install zeta-mlx-cli

Commands

# LLM Server (port 9044)
zeta-mlx llm start          # Start server
zeta-mlx llm start --daemon  # Start in background
zeta-mlx llm status         # Check status
zeta-mlx llm stop           # Stop server

# Embedding Server (port 9045)
zeta-mlx embedding start
zeta-mlx embedding status
zeta-mlx embedding stop

# Interactive chat
zeta-mlx chat

# Model management
zeta-mlx models list

API Usage

# Chat completion
curl -X POST http://localhost:9044/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model": "qwen3-8b", "messages": [{"role": "user", "content": "Hello!"}]}'

# Tokenize
curl -X POST http://localhost:9044/tokenize \
  -d '{"prompt": "Hello, world!"}'

Links

• GitHub
• Documentation