Use Models

After selecting a model in the Model Marketplace, you can view the model's endpoint information and usage examples, and copy the code directly to start calling.

Finding Model Endpoints

Go to the Model Marketplace, click on any model to enter its detail page, where you can see:

• Endpoint Management: Lists the available endpoint names, service providers, status, and billing information for that model
• Model Usage Examples: Provides curl and Python code examples that can be copied and used directly

The endpoint name format is {provider}@{model_name}@{interface_type}, for example:

• aliyun_bailian_bj@qwen/qwen3.7-max@open_ai — Alibaba Cloud Bailian-Beijing, OpenAI-compatible interface
• aws_bedrock_us_east_1@anthropic/claude-opus-4.6@anthropic — AWS Bedrock US-East-1, Anthropic interface

Two Interface Formats

AI Gateway supports two calling interfaces, depending on the protocol type of the model:

curl Calls

OpenAI-Compatible Interface (Qwen, etc.)

curl -X POST https://cn-shanghai-alicloud-aimesh.api.clickzetta.com/gateway/v1/chat/completions \ -H "Authorization: Bearer $API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "qwen/qwen3.7-max", "messages": [{"role": "user", "content": "hello"}] }'

Anthropic-Compatible Interface (Claude)

curl -X POST https://cn-shanghai-alicloud-aimesh.api.clickzetta.com/gateway/v1/messages \ -H "x-api-key: $API_KEY" \ -H "anthropic-version: 2023-06-01" \ -H "Content-Type: application/json" \ -d '{ "model": "anthropic/claude-opus-4.6", "messages": [{"role": "user", "content": "hello"}], "max_tokens": 256 }'

Python Calls

OpenAI-Compatible Interface (Recommended)

Use the openai library, pointing base_url to the AI Gateway endpoint:

from openai import OpenAI client = OpenAI( base_url="https://cn-shanghai-alicloud-aimesh.api.clickzetta.com/gateway/v1", api_key="<your-api-key>", ) response = client.chat.completions.create( model="qwen/qwen3.7-max", messages=[{"role": "user", "content": "hello"}], ) print(response.choices[0].message.content)

Anthropic-Compatible Interface (Claude)

Use the requests library to call the Anthropic Messages interface directly:

import requests url = "https://cn-shanghai-alicloud-aimesh.api.clickzetta.com/gateway/v1/messages" headers = { "x-api-key": "${API_KEY}", "anthropic-version": "2023-06-01", "Content-Type": "application/json", } payload = { "model": "anthropic/claude-opus-4.6", "messages": [{"role": "user", "content": "hello"}], "max_tokens": 256, } resp = requests.post(url, headers=headers, json=payload, timeout=60) print(resp.status_code) print(resp.text)

You can also use the official anthropic SDK, replacing base_url with the AI Gateway address:

import anthropic client = anthropic.Anthropic( base_url="https://cn-shanghai-alicloud-aimesh.api.clickzetta.com/gateway/v1", api_key="<your-api-key>", ) message = client.messages.create( model="anthropic/claude-opus-4.6", max_tokens=256, messages=[{"role": "user", "content": "hello"}], ) print(message.content)

Notes

• Endpoint addresses differ by deployment region; use the actual example code in the Model Marketplace as the reference
• OpenAI-compatible interface uses Authorization: Bearer <key>; Anthropic interface uses x-api-key: <key>
• Model names must exactly match the model identifier in the endpoint; copy from the Model Marketplace detail page
• API Keys must be created in advance in API KEY Management with calling permissions enabled for the corresponding model

Related Documents

• Quick Start — Create API Key and configure routing
• Model Pricing — Billing standards for each model
• Call LLMs in SQL — Calling via AI_COMPLETE in Lakehouse SQL