Make Your First Request

This guide uses Python 3.10 or later to call Chat Completions. Prepare a valid API Key, an available balance, and the complete model ID of a model that supports this API.

1. Set up Python

If Python is not installed, download it from the official Python website, then open a new terminal. Run the commands for your operating system below.

macOS / Linux:

python3 --version
mkdir -p certaedge-test
cd certaedge-test
python3 -m venv .venv
.venv/bin/python -m pip install --upgrade openai

Windows PowerShell:

py -3 --version
mkdir certaedge-test
cd certaedge-test
py -3 -m venv .venv
.\.venv\Scripts\python.exe -m pip install --upgrade openai

Later commands use the virtual environment's Python executable directly, so you do not need to run an activation script. If the directory already exists, enter it without creating it again.

2. Set your key and model

Replace sk-YOUR_API_KEY with the complete key you copied. Replace YOUR_MODEL_ID with a real ID from the model catalog. Do not send the placeholder values.

macOS / Linux:

export CERTAEDGE_API_KEY='sk-YOUR_API_KEY'
export CERTAEDGE_MODEL='YOUR_MODEL_ID'

Windows PowerShell:

$env:CERTAEDGE_API_KEY = 'sk-YOUR_API_KEY'
$env:CERTAEDGE_MODEL = 'YOUR_MODEL_ID'

These variables apply only to the current terminal and programs launched from it. Set them again when you open a new terminal.

3. Save and run the test program

Create first_request.py in the certaedge-test directory, and save the following code with a text editor. On Windows, make sure the file is not named first_request.py.txt.

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["CERTAEDGE_API_KEY"],
    base_url="https://certaedge.com/v1",
    timeout=60.0,
    max_retries=0,
)

response = client.chat.completions.create(
    model=os.environ["CERTAEDGE_MODEL"],
    messages=[{"role": "user", "content": "Introduce yourself in one sentence."}],
    stream=False,
)
print(response.choices[0].message.content)

Run it in the same terminal where you set the environment variables.

macOS / Linux:

.venv/bin/python first_request.py

Windows PowerShell:

.\.venv\Scripts\python.exe first_request.py

4. Confirm success

The terminal should print a model reply, and the request should appear in Usage Logs. Check the model, status, and usage before configuring your tools. This model request is billed at the selected model's rates.

Troubleshooting

Symptom	What to do
python3 or py not found	Install Python, open a new terminal, and check the version again.
No module named openai	Make sure you install and run with the same virtual environment's Python executable shown above.
KeyError: CERTAEDGE_API_KEY or CERTAEDGE_MODEL	Repeat step 2 in the current terminal.
401 / model not found	Check the key and model ID, then see Troubleshooting.
TLS, connection, or timeout error	Check the service address and network first. A failure before any HTTP response is not a model-name error.

If your model only supports Responses, Messages, or another protocol, this Chat Completions example does not apply. Choose a tool or request format that matches the supported protocol.