Kimi CLI - The Complete Command Line Guide

What is Kimi CLI?

Kimi CLI is a command line tool that brings the full power of Kimi API (by Moonshot AI / 月之暗面) to your terminal. Instead of switching to a web browser or GUI application, you can ask Kimi questions, generate code, analyze documents, and perform web searches entirely from the command line.

Built for developers and power users, Kimi CLI integrates seamlessly into existing workflows. Pipe file contents directly into prompts, script complex AI-assisted operations, and automate repetitive tasks with Kimi's 200K-token context window that handles even the largest codebases.

Whether you're debugging a tricky bug, generating boilerplate code, or summarizing lengthy documents, Kimi CLI is your terminal-based AI assistant that understands context and delivers accurate results.

Installation

Kimi CLI can be installed via npm or pip, depending on your preferred ecosystem. Choose the method that fits your workflow:

Install via npm

# Install globally with npm
npm install -g @moonshot/kimi-cli

# Or with pnpm
pnpm add -g @moonshot/kimi-cli

# Verify installation
kimi --version

Install via pip

# Install with pip
pip install kimi-cli

# Or with pipx for isolated environments
pipx install kimi-cli

# Verify installation
kimi --version

System Requirements

• -Node.js 18+ (for npm install) or Python 3.8+ (for pip install)
• -macOS, Linux, or Windows (WSL recommended)
• -A valid Kimi API key (get one at 40% off through our proxy)

Getting Started

After installation, configure your API key and start using Kimi CLI right away:

Configure Your API Key

# Set your API key (one-time setup)
kimi config set api_key YOUR_API_KEY

# Or use environment variable
export KIMI_API_KEY=YOUR_API_KEY

# If using our proxy, set the base URL too
kimi config set base_url https://kimi-api.com/v1

Basic Usage Examples

# Ask a simple question
kimi "What is the time complexity of quicksort?"

# Generate code
kimi "Write a Python function to merge two sorted arrays"

# Analyze a file
kimi "Explain this code" < main.py

# Pipe input for analysis
cat error.log | kimi "What's causing these errors?"

# Interactive chat mode
kimi chat

# Multi-turn conversation with context
kimi chat --model moonshot-v1-128k

Kimi CLI remembers conversation context within a session, so you can ask follow-up questions naturally, just as you would in a chat interface.

Key Features

Using Kimi CLI with Our Proxy - Save 60%

By pointing Kimi CLI at our proxy service, you get the exact same functionality at a fraction of the cost. Our proxy is fully compatible with the official Kimi API, so no code changes are needed beyond setting the base URL.

# Option 1: Set via CLI config
kimi config set base_url https://kimi-api.com/v1

# Option 2: Set via environment variable
export KIMI_BASE_URL=https://kimi-api.com/v1

# Option 3: Per-request override
kimi --base-url https://kimi-api.com/v1 "Your prompt here"

# Verify it's working
kimi "Hello, are you connected?" --verbose

Kimi CLI vs Alternatives

How does Kimi CLI compare to other AI command line tools? Here's a side-by-side comparison:

Related Resources