XycAi API Reference

XycAi speaks the OpenAI, Anthropic and Google Gemini wire formats. Point the official SDK at XycAi and call Claude / GPT / Gemini / GLM / DeepSeek / Kimi and 20+ other vendors without touching your business code.

• Base URL: https://cn.chatapi.app (primary; global / backup endpoints on the Endpoints page)
• Auth: API key (sk-…) created under API Keys in the console, routed by vendor by default
• Protocols: OpenAI-compatible (/v1/chat/completions), Anthropic native (/v1/messages), OpenAI Responses (/v1/responses, GPT models), Gemini native (/v1beta/models/{model}:generateContent), image generation (/v1/images/generations) and video generation (/v1/video/generations)

1. Base URL and headers

Pick the endpoint and the protocol you use; the widget shows the exact base URL to copy:

Each protocol authenticates with its own header:

2. Prerequisites

1. Sign up and log in

   Create an account at www.xycai.us.

2. Create an API key

   On the API Keys page create a key. By default it routes by vendor: pick Anthropic and the key can reach every Claude model, with the gateway choosing the cheapest available pricing group for each request. You can also restrict it by model or open it to all models and vendors. See API keys & routing.

3. Top up

   Requests are billed against your balance; every call appears in the usage log.

3. Your first request

The example sends a Chat Completions request to claude-sonnet-5-5 over the OpenAI-compatible protocol (which works for every vendor).

from openai import OpenAI

client = OpenAI(
    base_url="https://cn.chatapi.app/v1",
    api_key="sk-xxxxxxxx",
)

resp = client.chat.completions.create(
    model="claude-sonnet-5-5",
    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://cn.chatapi.app/v1",
  apiKey: "sk-xxxxxxxx",
});

const resp = await client.chat.completions.create({
  model: "claude-sonnet-5-5",
  messages: [{ role: "user", content: "Introduce yourself in one sentence." }],
});
console.log(resp.choices[0].message.content);

curl https://cn.chatapi.app/v1/chat/completions \
  -H "Authorization: Bearer sk-xxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5-5",
    "messages": [{"role": "user", "content": "Introduce yourself in one sentence."}]
  }'

Anthropic native protocol

With the Anthropic SDK or Claude Code, just point the base URL at XycAi:

from anthropic import Anthropic

client = Anthropic(base_url="https://cn.chatapi.app", api_key="sk-xxxxxxxx")
msg = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello, Claude"}],
)
print(msg.content[0].text)

curl https://cn.chatapi.app/v1/messages \
  -H "x-api-key: sk-xxxxxxxx" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-opus-5-5",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Hello, Claude"}]
  }'

4. Streaming

Add "stream": true to receive the response as Server-Sent Events. Use streaming for anything long-running (agents, long generations) so idle timeouts on intermediate network gear never cut the request.

stream = client.chat.completions.create(
    model="gpt-5.6-terra",
    messages=[{"role": "user", "content": "Write a four-line poem."}],
    stream=True,
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="", flush=True)

curl https://cn.chatapi.app/v1/chat/completions \
  -H "Authorization: Bearer sk-xxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-5.6-terra", "messages": [{"role": "user", "content": "Write a four-line poem."}], "stream": true}'

5. Popular models

Switch vendors by changing the model field. Frequently used names (full catalogue on the Models page):

6. Common errors

Failures return the structure below; the message ends with a request id — include it when contacting support:

{
  "error": {
    "code": "",
    "message": "Invalid token (request id: 2026102212xxxx)",
    "type": "new_api_error"
  }
}

Common status codes: 401 invalid key, 403 the key's routing mode does not include that model / vendor, 429 concurrency limit or insufficient balance, 400 bad request, 5xx upstream failure after automatic retries. See Errors.

7. Next steps