Manual setup
Use this page when a client is not listed or when you need the exact OpenAI-compatible values, smoke tests, and SDK examples.
Manual setup is the generic field list for any client that can target a custom Chat Completions Base URL. It is not a third-party product.
https://api.a.eus/v1Use these exact values
Most clients use the same five fields, even when the labels are slightly different.
Provider type
OpenAI-compatible / Custom OpenAI
Base URL
https://api.a.eus/v1
API key field
Your Asterionyx API key
Authorization header
Bearer YOUR_API_KEY
Model ID
Copy a current ID from the live Asterionyx Model Catalog
Smoke tests
Run these before configuring a larger app. Replace YOUR_API_KEY with the key from the portal.
curl https://api.a.eus/v1/models \
-H "Authorization: Bearer YOUR_API_KEY"curl https://api.a.eus/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "<MODEL_ID_FROM_CATALOG>",
"messages": [
{ "role": "user", "content": "Reply with one short sentence." }
]
}'SDK examples
Official OpenAI-compatible SDKs work when you change the API key and base URL.
from openai import OpenAI
import os
client = OpenAI(
api_key=os.environ["ASTERIONYX_API_KEY"],
base_url="https://api.a.eus/v1",
)
response = client.chat.completions.create(
model="<MODEL_ID_FROM_CATALOG>",
messages=[{"role": "user", "content": "Reply with one short sentence."}],
)
print(response.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.ASTERIONYX_API_KEY,
baseURL: "https://api.a.eus/v1",
});
const response = await client.chat.completions.create({
model: "<MODEL_ID_FROM_CATALOG>",
messages: [{ role: "user", content: "Reply with one short sentence." }],
});
console.log(response.choices[0]?.message?.content);Troubleshooting
401 / unauthorized
The API key is missing, pasted incorrectly, or belongs to another account. Copy it again from API Access.
402 / insufficient Credits
Authentication succeeded. HTTP 402 with error.code INSUFFICIENT_CREDITS. Create a recharge in the portal; do not replace the API key.
400 / unknown model
The model field is not currently callable. Copy a published ID from the live Model Catalog and try again.
No models in a dropdown
Open the live Model Catalog and type a published model ID manually.
Connection or 404 error
Confirm the Base URL includes /v1 unless the client explicitly asks for the host only.
Support needs a request reference
Share the JSON error.code and HTTP status. If there is no completion id, share the X-Asterionyx-Request-ID response header. Do not send your API key.
Choose a custom OpenAI-compatible provider
In your app, pick the provider option named OpenAI-compatible, custom OpenAI, custom endpoint, OpenAI proxy, or similar.
Use the Asterionyx Base URL
Paste the Base URL exactly as shown below. Keep the /v1 suffix unless the client explicitly asks for the host without a version path.
Paste your API key
Use the API key from the portal API Access page. It is sent as a standard bearer token.
Set a current Model ID
Copy a published Asterionyx model ID from the live Model Catalog and use it as the model name.
Run a small test
Ask for a one-sentence answer first. If that works, move the same Base URL, key, and Model ID into your app or workflow.