Documentation

The MAX API API is compatible with the OpenAI API. If you already use an OpenAI SDK, change two settings: the base URL and the API key.

Authenticate every request with your API key in the Authorization header. Keep your key secret — never expose it in browsers or mobile apps.

Authorization: Bearer YOUR_API_KEY

Send a list of messages and receive the model's reply. Any chat model ID from the catalog works.

curl https://clubs.byte-ai.cn/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.6-terra",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Set stream to true to receive tokens as server-sent events (SSE) while they are generated. The stream ends with data: [DONE].

curl https://clubs.byte-ai.cn/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.6-terra",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": true
  }'

Generate images with the Images API. Pick an image model from the catalog.

curl https://clubs.byte-ai.cn/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-1",
    "prompt": "A lighthouse on a cliff at sunset, watercolor",
    "size": "1024x1024"
  }'

Claude models can also be called with the native Anthropic Messages format at /v1/messages, using the x-api-key header (the Authorization header also works). This lets you use the official Anthropic SDK unchanged.

curl https://clubs.byte-ai.cn/v1/messages \
  -H "x-api-key: $API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Retrieve the models available to your key.

curl https://clubs.byte-ai.cn/v1/models \
  -H "Authorization: Bearer $API_KEY"

Errors use standard HTTP status codes and an OpenAI-style JSON body:

{
  "error": {
    "message": "Invalid API key",
    "type": "invalid_request_error",
    "code": "invalid_api_key"
  }
}

Default limits are generous and suitable for production workloads. If you exceed them you will receive 429 responses; retry with exponential backoff. Need higher throughput? Contact us through the console.

Long generations with reasoning models can take minutes. Set your client timeout to at least 300 seconds, or use streaming.