llktoken Developer Integration Guide

1 min read·OpenAI-compatible protocol

🌐 Language: English · 简体中文

Developer-facing API integration handbook · OpenAI-compatible gateway Service domain: https://global.llktoken.com API base URL: https://global.llktoken.com/v1

llktoken is an OpenAI-compatible relay gateway for large model APIs. Simply point your requests' baseURL at the llktoken address and swap in an API Key issued by llktoken, and you can access mainstream models such as GPT, Claude, Gemini, and DeepSeek through a single OpenAI-style calling convention — plus multimodal capabilities including images, audio, video, Midjourney image generation, and reranking.

1. Quick Start#

Get connected in three steps.

Step 1 · Create an API Key#

Sign in to the llktoken console, open the API Keys page, and create a key that starts with sk-. The plaintext key is displayed only once, at creation — copy and store it immediately (see Section 2).

Step 2 · Configure the baseURL and Key#

• Base URL: https://global.llktoken.com/v1
• API Key: your sk-... key

For most existing OpenAI integrations, you only need to replace these two values:

- baseURL: https://api.openai.com/v1
+ baseURL: https://global.llktoken.com/v1

- apiKey:  sk-你的原始Key
+ apiKey:  sk-你的llktoken-Key

Step 3 · Send your first request#

curl https://global.llktoken.com/v1/chat/completions \
  -H "Authorization: Bearer sk-你的llktoken-Key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [
      { "role": "user", "content": "你好，请用一句话介绍你自己" }
    ]
  }'

A chat.completion structure in the response means you are connected.

Gateway configuration: The public gateway is available at https://global.llktoken.com, and the Base URL is https://global.llktoken.com/v1. Request paths carry a single /v1 (e.g. /v1/chat/completions); writing /v1/v1/... returns 404.