OpenAI-Compatible API

Overview

The OpenAI Compatible API: Request Conclusion with OpenAI Schema operation returns results immediately without polling, using the standard OpenAI chat completions schema.

For general API information, see the embraceableAI OpenAI-Compatible API Documentation.

Parameters

Parameter	Required	Type	Description
Model	Yes	Dropdown (dynamically loaded)	Available models are automatically loaded based on your API key. Select from the dropdown.
Goal	Yes	Multi-line text	The primary objective or intent. Be specific about the desired outcome.
Details	Yes	Multi-line text	Context and information about the case. Provide relevant facts, constraints, and domain-specific information.
Policies	Yes	Collection or Text	Hard rules and constraints that must be followed. Choose between List Mode (separate list items, formatted with - when sent) or Text Mode (newline-separated text, each line = one policy, formatted with - ).
Stream Output Format	No	Boolean	Whether to return response in OpenAI stream format. Default: false. Note: n8n returns complete results after full generation, even when streaming is enabled.
Timeout	No	Number	Maximum time to wait for the API response (in seconds). Default: 300s (5 minutes), Range: 1-3600s. If exceeded, the node throws a timeout error.

Output Format

Returns response in OpenAI-compatible format:

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "model": "model-id",
  "choices": [{
    "message": {
      "role": "assistant",
      "content": "The complete reasoning process and final conclusion..."
    }
  }],
  "usage": {"prompt_tokens": 150, "completion_tokens": 200, "total_tokens": 350}
}

For detailed schema information and complete response structure, see the create-chat-completion endpoint documentation.

The reasoning process is in <think></think> tags within message.content, followed by the final conclusion. Review executions in the Sandbox Console under Logs.

n8n-Specific Features

• Immediate Response: No polling - returns results directly
• OpenAI Format: Compatible with OpenAI chat completion schema for integration with other tools
• Error Handling: Returns structured error responses for HTTP errors; throws timeout errors if timeout is exceeded

Differences from Conclusion API

Feature	Conclusion API	OpenAI-Compatible API
Response	Polling-based	Immediate
Format	Custom embraceableAI	OpenAI chat completion
Use Case	Long-running, full transparency	Quick responses, OpenAI compatibility

Use Conclusion API for long-running conclusions with automatic polling. Use OpenAI-Compatible API for immediate responses and OpenAI-compatible format.