Skip to content
WP EngineDocumentation

Getting started

This guide walks you from browsing the catalog to a working chat completion.

1. List available models

Section titled “1. List available models”

The catalog is public — no API key is required. Use it to discover model identifiers you will pass as the model field later.

Terminal window
curl https://api.ai.wpengine.com/v1/models

Truncated response:

{
  "object": "list",
  "data": [
    {
      "id": "google/gemini-3.5-flash",
      "object": "model",
      "owned_by": "google",
      "provider": "google",
      "display_name": "Gemini 3.5 Flash",
      "context_window": 1048576,
      "pricing": {
        "input_per_million": "1.5",
        "output_per_million": "9",
        "pricing_version": "2026-05-01",
        "unit": "credits"
      }
    }
  ]
}

Pick a model id from the response — you will pass it as the model field when you make a chat completion.

2. Create an API key

Section titled “2. Create an API key”

Create a key and copy the secret somewhere safe — it is shown once.

For this guide, an Account key with Full access is the quickest option; any key with Chat set to Write will do. Put it in an environment variable so the commands below can use it without pasting the secret:

Terminal window
export AI_API_KEY="wpe_xxx"

3. Make a chat completion

Section titled “3. Make a chat completion”
Terminal window
curl https://api.ai.wpengine.com/v1/chat/completions \
  -H "Authorization: Bearer $AI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-3.5-flash",
    "messages": [
      {"role": "system", "content": "You are concise."},
      {"role": "user", "content": "What is the speed of light in km/s?"}
    ]
  }'

Truncated response:

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1716240000,
  "model": "google/gemini-3.5-flash",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Approximately 299,792 km/s."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 15,
    "completion_tokens": 9,
    "completion_tokens_details": {
      "reasoning_tokens": 120
    },
    "total_tokens": 144
  }
}

On reasoning models, total_tokens also includes the tokens the model spent reasoning before it answered, reported separately in completion_tokens_details.reasoning_tokens, so it can be larger than prompt_tokens plus completion_tokens.

Streaming

Section titled “Streaming”

Set "stream": true in the request body to receive a stream of Server-Sent Events. Each event is a JSON chunk shaped like chat.completion.chunk; the stream terminates with data: [DONE].

See streaming a chat completion for a complete walkthrough.

Last updated: