Getting Started

This guide covers the shortest path from creating an API key to making a successful LLMAPI request.

Base URL

Examples in this documentation use:

https://api.llmapi.site

Use the public service URL shown by your deployment when it differs.

Create an API Key

1. Sign in to the website and open the API Key page.
2. Select Create Key.
3. Choose the group for the platform and models you want to call.
4. Configure quota, rate limits, and expiration when needed.
5. Submit the form, then copy and store the generated key.

See Create API Keys for group selection and verification details.

Configure Coding Clients

The website's Use Key action generates configuration for the selected key's group, public service URL, and supported client. Prefer the generated configuration over hard-coded examples.

Client	Requirement	Guide
Codex CLI, Codex Desktop, Codex IDE extension	A key whose group supports Responses API	Configure Codex
Claude Code and its IDE extension	A key whose Use Key dialog shows Claude Code	Configure Claude Code
CC Switch	A compatible key for the client you plan to configure	Configure CC Switch

Choose an Interface

Use case	Endpoint	Notes
Anthropic Messages	POST /v1/messages	Use a group that supports Messages.
Token counting	POST /v1/messages/count_tokens	Not available to OpenAI groups.
OpenAI Responses	POST /v1/responses	Recommended for Responses-compatible clients.
OpenAI Chat Completions	POST /v1/chat/completions	For general OpenAI-compatible clients.
List available models	GET /v1/models	Returns models available to the current key.
Query usage	GET /v1/usage	Returns quota mode and usage statistics.
OpenAI image generation	POST /v1/images/generations	Requires an OpenAI group with image generation enabled.
OpenAI image editing	POST /v1/images/edits	Supports multipart uploads and JSON image URLs.
Gemini native generation	POST /v1beta/models/{model}:generateContent	Requires a Gemini group.

Verify the Key

First list the models visible to the key:

curl https://api.llmapi.site/v1/models \
  -H 'Authorization: Bearer YOUR_API_KEY'

Then make a request with a model returned by that key. For example, an Anthropic-compatible group can call Messages:

curl https://api.llmapi.site/v1/messages \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "claude-sonnet-4-5",
    "max_tokens": 64,
    "messages": [{ "role": "user", "content": "Reply with OK." }]
  }'

Integration Notes

• API key group binding controls endpoint and model access.
• Use only models returned by GET /v1/models for the current key.
• OpenAI image endpoints require an OpenAI group with image generation enabled.
• Gemini image and multimodal requests use Gemini Native.
• Keep API keys out of source control, screenshots, logs, and client-side code shipped to browsers.