The flow
Starts it 1Model 4Logic 10Service 1
The steps in the order they run, from what starts it.
- 1When chat message receivedChat triggerStarts it
- 2OpenAI GPT 4.1Open AiModel
- 3Gemini 2.5 FlashGoogle GeminiModel
- 4Claude 4.5 SonnetAnthropicModel
- 5Preprocess OpenAI ResponseSetLogic
- 6Preprocess Gemini ResponseSetLogic
- 7Preprocess Anthropic ResponseSetLogic
- 8Combine responsesMergeLogic
- 9Add unit tests to responsesCodeLogic
- 10Iterate over each unit testsSplit In BatchesLogic
- 11Format Final ResultCodeLogic
- 12Wait for 3 secWaitLogic
- 13Final ResponseChatModel
- 14Run LMUnitContextual AiService
- 15Associate scores with ResponsesSetLogic
- 16Group Results TogetherCodeLogic
How it's built
- Starts on
- Someone opening a chat with it
- Steps
- 16, 2 loops or waits
- Size
- Medium: a working afternoon
- Connects to
- Nothing by connector; it calls services over plain HTTP
- Model
- Gemini, OpenAI
Questions
What does it need to run?
An n8n account, and a key for Gemini and OpenAI. You connect your own accounts after importing it; none are included.
What starts it?
It starts on someone opening a chat with it.
Does it use AI?
Yes. It calls Gemini and OpenAI. Model calls are billed by the provider, not by n8n.
How long does it take to set up?
Plan an afternoon. 16 steps is enough that you should read each one before running it.
How do I import it?
Copy the workflow, open a new workflow in n8n and paste. The steps appear on the canvas, ready for your own accounts.
Before you run itNothing here has been tested by us. It is somebody's working automation, read and described from its own steps. Treat the connections it wants as a list of accounts it will touch, and read the 16 steps before you point it at anything that matters.