Python: configure the client.
Install the openai package in your Python environment. Set CODIMIXA_API_KEY outside the source file and replace YOUR_MODEL_ID with a Chat Completions model from the catalog.
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["CODIMIXA_API_KEY"],
base_url="https://api.codimixa.com/v1",
)
result = client.chat.completions.create(
model="YOUR_MODEL_ID",
messages=[{"role": "user", "content": "Reply with one short greeting."}],
)
print(result.choices[0].message.content)JavaScript: use baseURL on the server.
Install the openai package in your Node.js project. Keep this code on the server, where the API key can stay in the environment.
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.CODIMIXA_API_KEY,
baseURL: "https://api.codimixa.com/v1",
});
const result = await client.chat.completions.create({
model: "YOUR_MODEL_ID",
messages: [{ role: "user", content: "Reply with one short greeting." }],
});
console.log(result.choices[0].message.content);Check the details that differ between SDKs.
Python uses base_url and api_key. JavaScript uses baseURL and apiKey. Use the base URL ending in /v1; the SDK method determines the request path.
Add features after the first request succeeds.
Check the model’s API format and supported features before adding streaming or tools. Inspect the error response if a feature call fails, and check your SDK’s retry settings so retries do not go unnoticed.
A FEW DETAILS
Before you continue
Why does changing a base URL still require a new key?
The key authenticates you to the service receiving the request. Use a Codimixa key with the Codimixa base URL.