Quickstart

This guide walks you through creating an API key and making your first request to the Kimchi Inference API.

Prerequisites

You need a Kimchi account. Sign up at app.kimchi.dev if you don't have one.

Steps

Create an API key

Go to Kimchi → API Keys and click + Create key. Copy the key and store it securely.

📘
The API Keys page has two tabs: API Keys for individual developer credentials, and Service Keys for automated workloads such as CI/CD pipelines. Both work identically as bearer tokens. See API Keys and Service Keys for details.

Make your first request

Send a request using cURL or any OpenAI-compatible client. Set your API key as an environment variable, then run:

export KIMCHI_API_KEY="your-api-key-here"

curl https://llm.kimchi.dev/openai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Accept: application/json" \
  -H "Authorization: Bearer $KIMCHI_API_KEY" \
  -X POST -d '{
    "model": "kimi-k3",
    "messages": [
      {
        "role": "user",
        "content": "Explain what a Kubernetes DaemonSet does in one sentence."
      }
    ]
  }'

The response follows the standard OpenAI chat completions format.

Configure your tool (optional)

The Kimchi CLI detects your installed AI coding tools and writes the correct config files automatically:

curl -fsSL https://github.com/getkimchi/kimchi/releases/latest/download/install.sh | bash

The interactive setup wizard will:

validate your API key
auto-detect installed tools (Claude Code, OpenCode, Cursor, Cline, Windsurf, Zed, and more)
write config files with the correct base URL and model list

For manual configuration, see the setup recipes for your specific tool.

What's next

Setup recipes

Step-by-step configuration guides for Claude Code, Cursor, OpenCode, Cline, Windsurf, and Continue.

Available models

See all models, their capabilities, and context window sizes.

Rate limits

Understand request limits and throughput.