Convika

Още не е на български, затова е на английски.

How do I use a copy-paste AI photo prompt in Gemini or ChatGPT with my own photo?

Quick answer

Copy the prompt, open Gemini or ChatGPT, add your photo and the prompt to the same message, and send. If the face changes, reply: "Keep my face exactly the same as in my photo. Change only the outfit and the background." PromptFoto shows each prompt with the picture it makes and opens Gemini or ChatGPT after you copy. Google's and OpenAI's help pages, read on November 19, 2026, give no number of free images: both free plans make images, and both tell you in the app when you reach the limit.

Publish my photo page free

Free plan: no credit card, 1 published page.

On this page

You saw a before and after on Reels: a plain selfie turned into a 1980s studio portrait, a couple on a motorbike at sunset, a saree portrait in golden light. The caption said "prompt in bio", and now you have the prompt on your clipboard. The gallery that gave it to you stopped there.

This guide covers the rest: where the prompt goes, how your photo gets into the same message, the line that keeps your face, what to do when the picture comes back wrong, and how many pictures the free plans make. The steps and limits come from Google's and OpenAI's own help pages, read on November 19, 2026, and each one is linked under Sources.

Use a copied photo prompt in 4 steps

1. Copy a prompt that shows its result

Pick the prompt by the picture it makes. On PromptFoto, every prompt sits beside the picture it makes. Tap Copy prompt, and a sheet opens with Open Gemini and Open ChatGPT, so the next step is one tap away.

2. Open Gemini or ChatGPT and add your photo

Use a clear, front-facing photo with good light.

  • Gemini, on a phone. In the Gemini app or at gemini.google.com, tap Add files, then Photos to pick from your photo library or Camera to take a new one. Google's help says you must be signed in.
  • Gemini, on a computer. At gemini.google.com, click Add files and upload the photo from your device.
  • ChatGPT, on a phone or a computer. OpenAI's help says: "Tap the + icon in the prompt area and select Add photos & files." On a computer, drag the photo into the message box or paste it from your clipboard. ChatGPT takes PNG, JPEG, and non-animated GIF files of 20 MB or less.

3. Paste the prompt into the same message and send

Press and hold in the message box, tap Paste, and send. The photo and the prompt go in one message, as Google's and OpenAI's steps both have it: the prompt says "my uploaded photo", and that is the photo attached to it. In Gemini the send button is Submit. OpenAI's help says a picture takes "a few minutes, depending on the complexity of your request", and you can keep using ChatGPT while it's made.

4. Save the picture

In the Gemini app, touch and hold the picture and tap Save. In Gemini on a computer, hover over the picture and click Download full size. In ChatGPT, open the picture and select Save to download it, or Share to send it to another app. Save with these buttons, not a screenshot: a screenshot is only as sharp as your screen.

Gemini and ChatGPT side by side

What each help page states, read on November 19, 2026.

GeminiChatGPT
Where you add the photoAdd files, then Photos or CameraThe + icon, then Add photos & files
Photos in one message10 files, 100 MB eachOpenAI gives no count: it "depends on various factors, including the size of the images"
Photo file types"Gemini Apps support most file types"PNG, JPEG, and non-animated GIF, 20 MB each
Sign-in"You must be signed in to Gemini Apps""On mobile web, you can also attach photos before signing in"
Who edits a photoAge 18 and over: "Editing images isn't available to users under 18""ChatGPT Images is available on all tiers." On a linked teen account, a parent sets it in Parental controls
Free planImage generation with Nano Banana 2 is included without a Google AI planFree includes "Limited image creation"
Number of free imagesGoogle doesn't sayOpenAI doesn't say
Aspect ratioGoogle's page doesn't say; write it in the prompt"Use the aspect ratio picker", or write it in the prompt
SavingDownload full sizeSave, or Share

The line that keeps your face

A good photo prompt opens with the face instruction. On PromptFoto's 1980s prompts, each prompt for one person starts: "Use my uploaded photo as the face reference and keep my face, skin tone and identity exactly the same." If the prompt you copied has no such line, put this one first:

Use my attached photo as the face reference. Keep my face, skin tone, and hairline exactly as in the photo.

If the picture still comes back with another face, don't start over. Reply in the same chat:

Keep my face exactly the same as in my photo. Change only the outfit and the background.

What else to change in a copied prompt

A prompt is a description in plain words, so edit the words:

  • Clothes. In PromptFoto's "80s hero with a scooter" prompt, swap "a wide-collar printed shirt" for what you'd wear: "a black kurta", "a denim jacket over a white t-shirt".
  • Background. Swap the place: in the same prompt, "a plain warm beige backdrop" becomes "a rooftop at sunset" or "a rainy street with neon signs".
  • Shape. For an Instagram post write "Vertical 4:5 portrait", for a story or WhatsApp status "Vertical 9:16", for a profile picture "Square 1:1". In ChatGPT, the aspect ratio picker does the same.
  • Who is in it. For a couple or a group, add one photo with everyone in it, or one photo per person, and say who is who: "The first photo is me, the second is my sister." PromptFoto's couple prompts are written for two faces. In Gemini without a Google AI plan, use one photo with everyone in it: Gemini picks the model for each prompt there, and Google's help says its image model "optimized for speed" "isn't designed for multiple image references or for multiple edits."

Here is a full prompt of ours to try:

Use my attached photo as the face reference. Keep my face, skin tone, and hairline exactly as in the photo. Show me as a 1970s rock guitarist on a small club stage: a brown suede jacket with fringe, a patterned shirt, and flared jeans, holding a sunburst electric guitar with no logo. Warm amber stage lights behind me, a little haze, faded colours, and film grain like a concert photo from 1975. Vertical 4:5 portrait, photorealistic, no text or watermark.

5 results that go wrong, and the fix for each

What you gotFix
Your face changedReply in the same chat with the keep-my-face line above. Use a brighter photo where your whole face shows. In Gemini, ask for one change per message
Extra fingers, or a stranger beside youName the part and ask again: "Redo this picture with 5 fingers on each hand and only me in the frame." In ChatGPT, open the picture, choose Select, highlight the hand, and describe the change
Gemini or ChatGPT refused, or the picture was removedIn Gemini, photo editing is for age 18 and over. Google removes a picture "when our systems detect a possible violation of Google's Terms of Service, including the Prohibited Use Policy", and tells you not to "violate others' copyright or privacy rights": use a photo of yourself, and of other people only with their OK, and take brand and celebrity names out of the prompt. In ChatGPT, a teen account linked to a parent follows the parent's setting in Parental controls
The picture is small or blurrySave it with Download full size in Gemini or Save in ChatGPT, not a screenshot. Start from a sharp photo: a blurred selfie gives the AI less of your face to keep
Letters on a sign or a shirt are garbledEnd the prompt with "no text or watermark". To get words, give them in quotes: a poster that says "Class of 2026". Google lists "reliable text rendering" for its Nano Banana 2 image model

How many pictures can you make for free?

Both free plans make images from your photo, and neither vendor's page gives a number. Read on November 19, 2026:

  • Gemini. Google's limits page lists image generation with Nano Banana 2 as included "Without a Google AI plan", at "Standard limits". Google AI Plus is "2x higher than standard limits" and Google AI Pro "4x higher". On the reset: "Your limit refreshes every 5 hours until you reach your weekly limit." Google doesn't say how many images the standard limit is. To see where you stand, open Settings, then Usage Limits at gemini.google.com.
  • ChatGPT. OpenAI's pricing page lists "Limited image creation" on Free, "More image creation" on Go, and "Unlimited and faster image creation" on Pro. The Free tier FAQ says: "File uploads, image generation, voice, data analysis, and other tools have separate usage limits. ChatGPT will notify you when you reach an applicable limit." OpenAI doesn't say how many images that is.

Two habits stretch a free limit. Google's page says long chats cost more usage: "Start a new chat when you move to a new topic." And pick the prompt by its picture before you send it, so every send is one you want.

Where to get prompts that show their result

PromptFoto is the gallery to copy from. Each prompt is shown with the picture it makes, so you choose the look before you spend a free image on it. The prompts are sorted by look and by occasion: 1980s, 90s, couple, family, birthday, wedding, saree, and headshot, and Trending this week collects the looks people are copying now. Copy prompt opens the sheet with Open Gemini and Open ChatGPT, and Share on WhatsApp sends a look to a friend. Every prompt is free to copy, and the site has pages in English, Indonesian, Portuguese, and Spanish.

Six photos sent one by one sink in a chat. One link shows them together as a page: an album, a birthday or wedding page, or the link in your Instagram bio. The AI that made the photos writes the page, and Convika gives it the link.

  1. Save the photos into one folder and name them photo1.jpg to photo6.jpg.

  2. Ask Gemini or ChatGPT for the page:

    Make one HTML file named index.html: a photo album titled "My 80s photos". Show photo1.jpg to photo6.jpg from the same folder in a grid, 2 across on a phone and 3 across on a computer. Put the CSS inside the file. No other files and no scripts.

  3. Save index.html into the same folder and drop the folder on Convika's publish tool. The box puts the photos inside the page. On November 19, 2026, we gave it a .zip of index.html and 6 JPEG photos of 1080 × 1350 px, 634 KB together, and the box answered "Combined index.html and 6 images into one file." The page came to 827 KB. On a phone, put the folder in a .zip and pick it with Choose a file.

  4. Click Get my preview link. You get a private link with no account, and it works for 3 minutes.

  5. Click Keep this page — free, no card and sign up. Click Publish page, choose Public under Who can open this page?, and click Next. The first time, click Publish free. The $5 button under it buys the Pro pass (30 days, paid once, no auto-renewal) and publishes without the Convika bar.

A page and its photos fit in 2 MB, which holds about 1.5 MB of photos. Save each photo as a JPEG about 1080 px wide: 6 of that size fit with room to spare.

Then send the link: share it on WhatsApp, or make it your link in bio with the link-in-bio template. If a photo shows as a broken icon, this guide finds the missing file. To choose the image model itself, see the best AI model for image generation.

Frequently asked questions

How do I use a photo prompt in Gemini?

Open the Gemini app or gemini.google.com and sign in. Tap Add files, then Photos, and pick your photo. Paste the prompt into the same message and tap Submit. Google's help page, read on November 19, 2026, says photo editing in Gemini is for age 18 and over.

How do I add my photo to ChatGPT with a prompt?

OpenAI's help says: "Tap the + icon in the prompt area and select Add photos & files." Pick your photo, paste the prompt into the same message, and send. On a computer, drag the photo into the message box. ChatGPT takes PNG, JPEG, and non-animated GIF files of 20 MB or less.

How do I keep my face the same in an AI photo?

Start the prompt with "Use my attached photo as the face reference. Keep my face, skin tone, and hairline exactly as in the photo." If the face still changes, reply in the same chat: "Keep my face exactly the same as in my photo. Change only the outfit and the background." Use a bright, front-facing photo.

Why did Gemini change my face?

The prompt didn't tell Gemini to keep it, or the photo was dark or turned away. Add the keep-my-face line and use a clear front-facing photo. For a couple or a group, give Gemini one photo with everyone in it: Google's help says its image model "optimized for speed" "isn't designed for multiple image references or for multiple edits."

Are Gemini AI photo prompts free?

Yes. A prompt is text you paste, and Gemini makes images without a Google AI plan: Google's limits page lists image generation with Nano Banana 2 under "Without a Google AI plan" at "Standard limits". Google's page, read on November 19, 2026, gives no number of images. Every prompt on PromptFoto is free to copy.

How many images can I make for free in ChatGPT?

OpenAI doesn't give a number. Its pricing page lists "Limited image creation" on the Free plan, and its Free tier FAQ says: "ChatGPT will notify you when you reach an applicable limit." Read on November 19, 2026.

Do the same photo prompts work in Gemini and ChatGPT?

Yes. A photo prompt is plain text, and both apps take a photo and a prompt in one message and return a picture. Paste the same prompt into either one. In ChatGPT, the aspect ratio picker sets the shape; in Gemini, write the shape in the prompt.

Ask Gemini or ChatGPT for one index.html album page that shows your photos, put the page and the photos in one folder, and drop the folder on Convika's publish tool. The box puts the photos inside the page and gives you a link to send on WhatsApp or put in your Instagram bio.

Sources

Each page was read on November 19, 2026, between 15:00 and 16:05 Japan time.

Next step

How to password protect an HTML page or website

A JavaScript password script leaves the password in the page source. To password protect an HTML page or link, encrypt the file or let a server check it.

Related articles

More ways to send your page

Skipbox

Cheap & fast GitHub Actions runners, unlimited for $39 a month

Your own machine keeps Docker layers and package caches between jobs. Switch by changing one runs-on line.

Start a free Skipbox trial⁠

Share this article

Put your AI photos on one link

Ask your AI for one album page, drop the page and its photos in, and send one link on WhatsApp or in your Instagram bio.

Publish my photo page free

Free plan: no credit card, 1 published page.