Quickstart

XycAi is a unified AI API gateway that exposes dozens of upstream model providers through a single OpenAI-compatible interface. If you have used the OpenAI SDK before, you barely need to change any code — just swap the base_url and api_key.

Prerequisites

1. Sign up and log in to the console.
2. Create a key on the API Keys page; it looks like sk-xxxxxxxx.
3. Make sure your account has available balance.

Base URL

All requests go to the same gateway domain. Depending on the SDK or how you call it, the base_url takes one of three forms:

Authentication

Pass your key in the request header:

Authorization: Bearer sk-xxxxxxxx

A few native protocols use different headers: the Claude-native /v1/messages uses x-api-key, and the Gemini-native /v1beta uses x-goog-api-key. See Authentication.

Your First Request

The example below sends a chat completion with claude-sonnet-4-6.

from openai import OpenAI

client = OpenAI(
    base_url="https://apicdn.xyc.ai/v1",
    api_key="sk-xxxxxxxx",
)

resp = client.chat.completions.create(
    model="claude-sonnet-4-6",
    messages=[
        {"role": "user", "content": "Introduce yourself in one sentence."},
    ],
)

print(resp.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://apicdn.xyc.ai/v1",
  apiKey: "sk-xxxxxxxx",
});

const resp = await client.chat.completions.create({
  model: "claude-sonnet-4-6",
  messages: [
    { role: "user", content: "Introduce yourself in one sentence." },
  ],
});

console.log(resp.choices[0].message.content);

curl https://apicdn.xyc.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-xxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "messages": [
      {"role": "user", "content": "Introduce yourself in one sentence."}
    ]
  }'

Response Structure

Returns a standard OpenAI-compatible structure:

{
  "id": "chatcmpl-xxxxxxxx",
  "object": "chat.completion",
  "created": 1748600000,
  "model": "claude-sonnet-4-6",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "I am an AI assistant ..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 18,
    "total_tokens": 30
  }
}

Streaming

Add "stream": true to receive the response chunk by chunk (Server-Sent Events).

stream = client.chat.completions.create(
    model="gpt-5.4",
    messages=[{"role": "user", "content": "Write a four-line poem."}],
    stream=True,
)

for chunk in stream:
    delta = chunk.choices[0].delta.content or ""
    print(delta, end="", flush=True)

curl https://apicdn.xyc.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-xxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.4",
    "messages": [{"role": "user", "content": "Write a four-line poem."}],
    "stream": true
  }'

Popular Models

The same endpoint can switch between models from different providers — just change the model field. A few popular ones:

See the full list at Models.

Common Errors

On failure the response has the structure below, with a request id appended to message to aid debugging:

{
  "error": {
    "code": "",
    "message": "Invalid token (request id: 2026053012xxxx)",
    "type": "new_api_error"
  }
}

Common HTTP status codes: 401 unauthenticated, 403 forbidden, 429 rate limited / insufficient balance, 400 bad request, 500/503 upstream error. See Errors.

Next Steps