ChatGPT
ChatGPT is for calling OpenAI models inside your scenario: text, images, and files. Use it for chat, document analysis, and structured replies without your own OpenAI API key.
Plug and Play (PnP) billing
This is a PnP (Plug and Play) node: Latenode meters usage and charges PnP tokens on top of scenario CPU seconds (1 PnP token = $1). You do not add your own API key on this node. Per-model prices are shown in the node settings where you pick the model.
Your own API key
To use your provider account and key instead, add the Custom LLM Connection node and a connection with the provider Base URL (the key is stored in the connection).
Getting structured output
Default replies are plain text.
For JSON that downstream nodes can parse (extract fields, classify, fill forms), enable Structured Output and describe the JSON shape in the prompt.
Example (receipt):
You are given a receipt text. Extract the store name, purchase date, list of items with prices, and the total amount.
Receipt:
{receipt_text}
Respond in JSON using this format:
{
"store": "...",
"date": "...",
"items": [
{ "name": "...", "price": 0.00 }
],
"total": 0.00
}
Advanced: Output JSON Schema
For a strict field structure, fill in Output JSON Schema.
Schema format
ChatGPT and OpenRouter use a name wrapper around the schema. Claude starts with "type": "object" instead. Copying between nodes without reformatting can fail.
{
"name": "receipt_data",
"schema": {
"type": "object",
"properties": {
"store": { "type": "string", "description": "Store name" },
"date": { "type": "string", "description": "Purchase date" },
"total": { "type": "number", "description": "Total amount" }
},
"required": ["store", "date", "total"],
"additionalProperties": false
},
"strict": true
}Fields
Google AI
Gemini, Gemma, Imagen, Nano Banana, Google AI Mode, and autocomplete, PnP billing.
OpenAI Image Generation
Generate and edit images with GPT Image / DALL-E, PnP billing.
Need Help? Ask the community
If something on this page is missing or unclear, post on the Latenode community forum. Our team and other users usually reply quickly.