Generate ad image variations using GPT-4, Dumpling AI & Google Drive
Needs changesNeeds changes. This template uses 1 thing(s) that n8n 3.0 removes: AI Agent v1 (old agent modes). 1 breaking item(s) found. Static check, last run 2026-10-19.
What this workflow does
This AI workflow has 13 nodes and runs when someone submits an n8n form (“Ad Image Generator”). Along the way it connects Google Drive, OpenAI, Google Sheets and OpenAI Chat Model. It also calls app.dumplingai.com over HTTP.
How it runs
Trigger: When someone submits an n8n form (“Ad Image Generator”).
- Upload Ad Image to Google Drive (Google Drive): uses Google Drive
- Download Ad Image for Analysis (Google Drive): download
- Describe Visual Style of Image (OpenAI): analyze (image)
- Analyze Brand Website Style (OpenAI): uses OpenAI
- LangChain Agent: Generate Variation Prompts (AI Agent): runs an AI agent, with model: GPT-4o (Connected to LangChain Agent); output parser: Parse Prompts into JSON Array
- Split: One Prompt per Item (Split Out): splits a list into separate items
- Download Base Image for Each Variation (Google Drive): download
- Loop: Process Image Variations (Loop Over Items (Split in Batches)): loops over the items in batches
- Dumpling AI: Generate Image Variation (HTTP Request): calls app.dumplingai.com (POST)
- Log Image Variation URLs to Google Sheets (Google Sheets): append
Services it uses
HTTP calls to: app.dumplingai.com
Before you upgrade to n8n 3.0
1 thing in this workflow is removed in n8n 3.0. Fix it before you upgrade:
- AI Agent v1 (old agent modes) in LangChain Agent: Generate Variation Prompts (v1.7): replace with Latest AI Agent node (Tools Agent). For SQL Agent, use a Postgres or MySQL tool with a recent AI Agent.
Rather not rewrite it? Many of these flows can be rebuilt in Make (free plan).
| Removed in 3.0 | Replace with |
|---|---|
| AI Agent v1 (old agent modes) LangChain Agent: Generate Variation Prompts (v1.7) | Latest AI Agent node (Tools Agent). For SQL Agent, use a Postgres or MySQL tool with a recent AI Agent. |
Nodes used (10)
- n8n Form Trigger
n8n-nodes-base.formTrigger - Google Drive
n8n-nodes-base.googleDrive - OpenAI
@n8n/n8n-nodes-langchain.openAi - AI Agent
@n8n/n8n-nodes-langchain.agent - Structured Output Parser
@n8n/n8n-nodes-langchain.outputParserStructured - Split Out
n8n-nodes-base.splitOut - Loop Over Items (Split in Batches)
n8n-nodes-base.splitInBatches - HTTP Request
n8n-nodes-base.httpRequest - Google Sheets
n8n-nodes-base.googleSheets - OpenAI Chat Model
@n8n/n8n-nodes-langchain.lmChatOpenAi
From the author
🧾 What this workflow does This workflow takes a reference ad image and brand website, then uses GPT-4, LangChain, and Dumpling AI to generate 10 high-quality image variations for ad testing. These image variations are visually consistent but subtly different…