
Give every person a number and a job
Write "image 1 is Mum, image 2 is Dad, image 3 and 4 are the children". There is no role selector, so the prompt is where each photo gets its place.
Upload a photo of each person, describe the setting, and get one warm family portrait. Useful when everyone is in a different place.
Make one family portrait from separate photos. Upload up to five people, pick the setting, and keep each face. Good for holiday cards. First run free.
Upload at least one reference image before generating.
Updated
All four accept more than one photo at once. Keeping five faces distinct is the hard part, so run a larger group through two models.

OpenAI latest image model. Sharper instruction following, cleaner in-image text, up to 4K.

Photorealistic detail with tight prompt control. Accepts up to 8 reference images.

Highest quality Google model. Supports up to 4K resolution.

Strong typography and poster layouts. Handles dense in-image text at 1K and 2K.
Families are scattered and calendars never line up. These points make the combined picture look like a real moment.

Write "image 1 is Mum, image 2 is Dad, image 3 and 4 are the children". There is no role selector, so the prompt is where each photo gets its place.

A sharp, evenly lit face per person works best. Sunglasses, hats, or a face in profile make it hard to keep a likeness.

Add "keep each person’s face and age exactly as in their own photo". Without it, similar faces blur together and ages drift.

A single scene, such as "a sofa by the fireplace, soft evening light", reads as one moment. Photos taken in different light look pasted unless you ask for one shared light.

Searches for family portraits climb from November. Generate in 3:2 or 4:3 for cards, and check each face at full size before ordering prints.

Beyond five people, start from a group generated here and use it as one reference, adding the rest in a second pass.

The picture is redrawn, not a real photo of the group. Be clear about that if you share it, and only use photos of people who have agreed.
Same generator, pointed at a neighbouring job.
The general version: merge people, products, and style from up to five photos.
Add one missing family member to a photo you already have.
Just two people? Pair one photo of each into a couple shot.
Dress everyone in matching outfits for the card.
The general reference-image workspace for edits of any kind.
Upgrade for larger monthly credits, faster generation, and commercial-ready output.
Try GPT-IMG with signup credits
For regular creators and light production use
Generate 50–125 images / month
For high-volume image generation workflows
Generate 120–300 images / month
Buy credits once and use them whenever you need. Purchased credits never expire.
Practical details for this generator page.
Upload a clear photo of each person, number them in the prompt, describe one shared setting and light, and ask to keep each face unchanged. Generate and check every face.
Up to five photos per generation. For a bigger group, make a first portrait here, then use it as one of the references for a second pass.
Mostly, with clear photos and the face-lock clause. The frame is redrawn, so compare each face with its own photo before you print or send it.
Yes, upload their photo as one of the references and say where they should stand. Only use photos of people who have agreed to it.
New accounts start with 30 signup credits. A 1K Standard edit on GPT Image 2.5 is 15 credits, so your first two family portrait runs are covered. Images made on free signup credits carry a watermark.
Start with GPT Image 2.5, the default. FLUX 2 Pro costs the same 15 credits for a 1K Standard run, so you can compare the two on one brief. Group likeness is the hard part, so run the same set through two models and keep the one where the faces stay distinct.
Use the prompt panel above with the same account, credits, and history workflow.