Connect to your dedicated endpoint

Note

This page covers Dedicated Capacity. Models listed here are not available with an Inference API key unless they also appear in the Inference API catalog.

Base URL and API keys

Your deployment has its own base URL and its own API keys. We give you both at onboarding.

The examples on this page use the placeholder https://<your-endpoint>/v1. Replace it with the base URL that we give you.

Make a request

The API is the same OpenAI-compatible API as on the Inference API. Change the base URL and the API key. The rest of your code stays the same.

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://<your-endpoint>/v1",
    api_key="your-dedicated-api-key"
)

completion = client.chat.completions.create(
  model="GLM-5.3",
  messages = [
      {"role": "system", "content": "Answer the question in a couple sentences."},
      {"role": "user", "content": "Share a happy story with me"}
    ]
)

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

cURL

export API_KEY="your-dedicated-api-key"
export URL="https://<your-endpoint>/v1/chat/completions"

curl -H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"messages": [
{"role": "system", "content": "Answer the question in a couple sentences."},
{"role": "user", "content": "Share a happy story with me"}
],
"model": "GLM-5.3"
}' \
-X POST $URL

Use the model IDs of your deployment. See Supported models.

Features

These guides apply unchanged. Use your own base URL and a model of your deployment.

• Text generation
• Function calling and JSON mode, including structured output
• Vision
• Reasoning
• API reference

A feature works only on a model that supports it. For example, vision needs a model that accepts images.

Rate limits

The rate limits of your deployment are agreed with you. They are not set by tier. The tier tables on Rate limits apply to the Inference API only.