n8n

n8n can call sivcloud because its API is OpenAI-compatible. You need an API key from the API keys page and a model id from the Models page in the console.

OpenAI credential

1. In n8n, create a new credential of type OpenAI.
2. Put your sivcloud API key in API Key.
3. Set the base URL to https://ai.sivcloud.com/v1. If your n8n version shows no base URL field in the credential, look for it in the node options; its name can differ between versions.
4. Save, then run the connection test if there is one. The n8n test calls the model list; if it fails but the chat node works, ignore it.

OpenAI Chat Model node

1. Add the OpenAI Chat Model node as a sub-node of an AI Agent or a Basic LLM Chain.
2. Pick the sivcloud credential.
3. Type the model id from the Models page, for example deepseek-v4.1-flash. The built-in n8n list is OpenAI’s own, so a sivcloud model id has to be typed.
4. Leave Use Responses API off so n8n uses chat completions.

n8n’s built-in OpenAI tools such as web search and file search are not available through sivcloud.

Another way: the HTTP Request node

If the node above does not fit, send the request yourself:

• Method POST, URL https://ai.sivcloud.com/v1/chat/completions.
• Header Authorization set to Bearer <your API key>, and Content-Type set to application/json.
• JSON body:

{
  "model": "deepseek-v4.1-flash",
  "messages": [{ "role": "user", "content": "Hello" }]
}

The full format is on the endpoint page.

Common errors

• 401: the API key is wrong, blocked or expired. Check the key in the console.
• 402: the balance is empty. Add balance on the Balance page.
• 404: the model id does not exist. Copy the id exactly from the Models page.

The full list is on Error codes. Do not put the API key in a workflow you share; keep it in an n8n credential.