Generate Images: ChatGPT + Make

A board-game designer needs forty pieces of placeholder card art for playtesting: goblins, potions, traps, the lot. Generating them one at a time in a chat window, downloading each, renaming each, is an evening lost. The ideas are already written in a spreadsheet; the tedium is everything after.

This build does everything after. You keep a sheet with one image idea per row. Make sends each to ChatGPT’s image model, gets the picture back, and saves it to a Google Drive folder, named and tidy. A list of ideas becomes a folder of images while you do something else.

We wire it in Make, the first build in our AI image and vision set. This is the corner where Claude steps aside: it does not generate images, so the right tool here is ChatGPT (OpenAI). Make is what turns a one-at-a-time chat task into a batch that runs itself.

What does this workflow actually do?

In one line: a spreadsheet row becomes a saved image. Make watches your sheet. When you add a row with an image prompt, it sends the prompt to OpenAI’s image model, receives the generated picture, and uploads it to a Google Drive folder. No chat window, no downloading, no renaming.

A few real cases, none of them the usual ones:

• The board-game designer above, batch-generating placeholder art for a playtest.
• An Etsy candle maker generating seasonal product mockups from a list of scents.
• A tabletop dungeon master generating monster portraits for a whole campaign at once.
• A preschool teacher generating coloring-page line art for each week’s theme.

Writing the ideas is the creative part, and you still do that. Generating, saving, and naming each one is the busywork. That busywork is what the automation removes.

Why use Make if ChatGPT can generate images itself?

Fair question. The ChatGPT app makes one image when you ask, in a window, and then you download and file it by hand. That is perfect for a single picture. It is painful for forty.

Make turns the one-at-a-time task into a batch. It watches the sheet, sends every row to the image model, and files each result with no clicking. The model still does the creative step, the actual image; Make does the repetitive plumbing of doing it many times and putting the results where they belong. The tool and the model split the work, and that split is what makes volume painless.

What do you need before you start?

• A Make account. The free 1,000 operations a month covers plenty of images.
• An OpenAI API key from the OpenAI platform (the developer key, not the chat app). New to AI tools? Our how to use Claude guide explains API keys in general.
• A Google account with a Drive folder to save the images into.
• A Google Sheet with one column, Prompt, where each row is an image idea.

Two Make words, once. A scenario is one automation, the whole sheet-to-Drive chain. A module is one box in it. Three boxes, one scenario.

How does the workflow work, step by step?

Three modules, left to right:

Module	App	What it does
1. Trigger	Google Sheets	Fires when you add a row with an image prompt
2. Brain	OpenAI	Generates the image from the prompt
3. Output	Google Drive	Uploads and saves the image file

Step 1: Set up the prompt sheet

Create a scenario in Make and add Google Sheets, Watch New Rows. Connect your account and point it at a sheet with a Prompt column. Each new row you add is one image request. The trigger hands the prompt to the next step.

Step 2: Let OpenAI generate the image

Add the OpenAI, Generate images module, paste your OpenAI API key, and map the Prompt column from Step 1 into the prompt field. Pick a size and how many images per prompt. A good prompt is specific about subject and style:

A friendly cartoon goblin holding a lantern, flat illustration
style, soft colors, plain white background, centered.

Adding the same style words to every row, flat illustration, soft colors, is what keeps a whole set looking like it belongs together.

Step 3: Save the image to Drive

Add Google Drive, Upload a File, connect your account, choose the folder, and map the image data from Step 2 as the file. Name it from the row so the folder stays organized. Now every prompt you add becomes a saved, named image.

How do you run it and check the result?

Click Run once after adding a test prompt. When I tested this, a sheet of prompts turned into a folder of images in a couple of minutes, and the fix that mattered was adding consistent style words to each prompt so the set looked unified instead of random. Tune your prompts, then turn the scenario on so each new row generates its image.

After that, image ideas stop being a chore and become a list. You write what you want, and the folder fills in the background, ready when you need it.

What does this cost to run?

Piece	Free tier	If you outgrow it
Make	1,000 operations/month free	Core plan from about $9/month
OpenAI images	Pay per image	Standard images run a few cents each, more than text
Google Drive	Free	Free with your account
Google Sheets	Free	Free

Images cost more than text generation, a few cents each rather than a fraction of a cent, so watch your volume. Otherwise the build sits inside the free Make plan. See tiers in our Make guide.

What can go wrong, and how do you avoid it?

• The bill adds up. Images cost cents each, not fractions of a cent. Generate in batches you actually need, not thousands on a whim.
• The set looks inconsistent. Add the same style words to every prompt so the images share a look.
• A prompt is refused. Image models block certain content; rephrase, and keep prompts clean.
• Files overwrite each other. Name each upload from the row so two images never collide.

Want more of this pattern? Our Make AI scenarios roundup has the next ideas.

How do you build this in Zapier or n8n instead?

The three jobs do not change, only the names do.

Job	Make	Zapier	n8n
Catch a prompt	Watch New Rows	New Spreadsheet Row trigger	Google Sheets Trigger node
Generate the image	OpenAI Generate images	OpenAI (DALL-E) action	OpenAI node
Save the file	Upload a File	Google Drive Upload	Google Drive node

Make and Zapier are the gentlest for a first build. Zapier vs Make vs n8n compares all three.

Common questions

Why not just use the ChatGPT app?

For one image, do. This build is for volume: a list of prompts becomes a folder of saved, named images with no clicking, which the app cannot do on its own.

Can I use a different image model?

Yes. Swap the OpenAI module for Stability, Gemini, or another image app. The sheet trigger and Drive upload do not change.

How do I keep a consistent style?

Put the same style words in every prompt, or add them once in the module. Consistency comes from the words, not the model.

Can it post images straight to social or a site?

Yes. Swap or add a final module for WordPress, Buffer, or similar. Saving to Drive first keeps it simple to check.

Does Claude do images?

No, which is why this build uses OpenAI. Claude is for reading and writing; image generation needs an image model.

Sources and docs

• Creating a scenario (Make Help)
• OpenAI image generation (docs)
• Upload files to Google Drive (Google)
• OpenAI on Make (integrations)
• DALL-E — Grokipedia

Last reviewed: May 2026. Make, OpenAI, and Google update their interfaces; check the official pages for exact button names.

You may also like

• AI Automation for Beginners (the hub)
• Draft WordPress posts from a spreadsheet with Claude
• Summarize new documents with Claude + Make
• Make.com Complete Beginner’s Guide
• How to use Claude AI
• Zapier vs Make vs n8n: which to start with
• The AI Tools Directory