One selfie in · Retro class portrait out · Free, no app
Big hair, a laser backdrop, faded film stock. Your face stays yours, you pick the era, and the print-style shot comes back in seconds.
A real class-portrait render made here, on the model this page runs. Drag the divider to compare the modern selfie with the retro result. Every output is AI-generated, so your finished picture varies with your photo, your prompt, and the decade you describe — treat this as a preview of what the tool can do, not a fixed result.


Drag to compare before and after
The mottled blue laser backdrop, the soft-focus glow, the feathered hair, the collared shirt under a sweater — one selfie restyled as the class portrait you would have sat for in 1996, down to the faded colour of a print that has been in a drawer for thirty years.
Instead of digging through a shoebox or booking a novelty studio, you upload a clear picture, describe the decade you want, and the model rebuilds the shot while keeping your face and expression recognizable. On Kavel the tool runs on the Nano Banana image model, so a finished throwback arrives in one generation rather than an afternoon of manual filters.
You start from your uploaded photo — the flow defaults to image-to-image — set the format and the resolution, and the credit estimate updates before you commit; a failed run is refunded automatically (unless it broke the content policy), so trying a second decade costs nothing but time. The result suits a group chat that will not stop laughing, a creator riding a nostalgia trend, or anyone who wants to see what they would have looked like in their parents' high school.
And because the same studio also ages a face, swaps outfits, and repairs old prints, one workflow covers a range of looks without a new tool for each. Think of it less as a filter and more as a time machine with a laser backdrop — with the direction in your hands, and your own face in the frame.
Creative engine
Upload a selfie, describe the decade, choose the format and resolution, and the credit estimate updates before you generate.
Upload one clear selfie and the model reads your face, then rebuilds the shot as a dated school portrait.
Describe the decade you want — grunge nineties, glamour-shot soft focus, an eighties throwback — and the styling follows that brief.
Set the format, check the credit estimate, and generate; a run that fails costs you nothing, so a second era is free to try. A yearbook portrait fixes the era at about 1994. For something you send rather than file away, the Christmas card and the movie poster use the same photo with a different job.
From one selfie to a dated school portrait, with the cost shown before you commit.
Add one clear photo with your face and shoulders visible. Even light and a plain background give the model the most to restyle.
Set the era: a mottled laser backdrop, feathered hair, a sweater over a collar, soft-focus glow, faded print colour. Specific direction keeps the styling on-brief.
Confirm the estimate, generate the portrait, and download it. If a run fails, it is free, so you can adjust and try another decade.
Reunions, group chats, birthday slideshows — the same retro treatment, different excuses.
Drop a retro class portrait of yourself into the group thread and watch it get quoted back for a week.
Ride a throwback trend with a whole set of dated portraits, made from photos you already have.
Print a fake class portrait for a reunion, a birthday card, or a gag gift that lands harder than a real photo would.
Common questions about turning a selfie into a retro class portrait with Kavel.
The backdrop becomes the mottled laser sheet every school used, the light goes soft and hazy, the hair gets feathered, the clothes turn into a collar under a sweater, and the colour fades like a print that sat in a drawer. Your face carries through it all.
No. Free credits arrive with the account, and a run that fails is refunded automatically (unless it broke the content policy). Beyond that, cost tracks the resolution and the number of images, and the estimate appears before you generate.
A clear, front-facing shot with your face and shoulders in frame. The joke only lands if it is unmistakably you under the dated hair, and that needs a sharp starting photo.
Yes. Early-nineties grunge, a late-nineties glamour shot, an eighties throwback with a heavier soft focus. Each has its own hair, its own clothes, and its own backdrop, so name the one you want.
Because that is the era. The feathered, over-blow-dried look is doing most of the work — without it the picture just looks like a photo on a blue background. If you want your own hair kept, say so, though the result will read less like the real thing.
You can run each person separately and lay the results out as a page, which is how most people use it. A single photo with several faces in it is harder, because the styling has to land on everyone at once. The era cues do most of the work: a soft-focus vignette, a mottled blue-grey backdrop, feathered layers, and the slightly warm colour cast of film that sat in a drugstore envelope for a week before anyone saw it.
Yes — that is a different ask and worth naming precisely. A yearbook portrait means the studio setup: mottled blue backdrop, shoulders squared to camera, even key light. A 35mm restyle keeps whatever composition your photo already has and changes the film stock instead: visible grain, slightly cool shadows, highlights that roll off rather than clip. Say which of the two you want, because asking vaguely for a film look tends to land halfway between them and read as neither.
Three things, in the order they matter: grain that sits in the midtones rather than the shadows, a colour cast that shifts the whole frame instead of just the highlights, and shadows that stay open rather than crushing to black. Naming those beats asking for vintage — vintage is a mood, and the model will pick whichever decade it feels like.
Usually because the original was shot in flat, even light. Film's character shows up in how it handles contrast, so a photo with no strong light direction gives the restyle nothing to work with — the grain lands on top of an already-flat image and reads as noise. A picture with a window, a lamp, or sun on one side of the face converts far more convincingly.
One credit pool covers Nano Banana images and Seedance and Veo video. The cost shows before every run, and failed jobs are not charged. Use a subscription for ongoing work, or a one-time pack when you just need to top up.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
One-time top-ups — buy extra credits any time you run low.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
Charges appear as “KAVEL AI” on your card statement. You can cancel any time from Settings → Billing; cancellation takes effect at the end of the period you already paid for.
Powered by
Use one balance across every supported image and video model — eleven of them today, listed below — and check the credit cost before each request.
Video models
Image models
Model credit guide
You're only charged for successful generations. The exact estimate in the generator varies by model, length, resolution, audio, and number of images.
| Type | Model | Credit cost |
|---|---|---|
| Video | Seedance 2.5 | The long-take tier, priced per second: 5s 480p ≈ 211 credits, 5s 720p ≈ 473. A full 30s take runs ≈ 1,260 at 480p and ≈ 2,835 at 720p. Supplying a reference clip lowers the per-second rate. |
| Video | Seedance 2.0 | 5s 720p image-to-video ≈ 188 credits; text-to-video ≈ 308 credits. Scales with resolution and length. |
| Video | Seedance 2 Fast | Faster and lower cost. 5s 720p text-to-video ≈ 248 credits. |
| Video | Seedance 2 Mini | The cheapest Seedance tier. 5s 720p ≈ 154 credits, 5s 480p ≈ 72 credits. |
| Video | Seedance 1.5 Pro | Audio doubles the rate. 5s 720p ≈ 27 credits silent, ≈ 53 with audio; 1080p ≈ 57 and ≈ 113. |
| Video | Veo 3.1 | Billed per video, not per second (Lite tier). About 45 credits at 720p and 53 at 1080p. |
| Video | Kling 3.0 | Audio raises the rate. 5s 720p ≈ 105 credits silent, ≈ 150 with audio; 1080p ≈ 135 and ≈ 203. |
| Video | MiniMax H3 | Fixed 2K, no resolution ladder. Priced per second — a 5s clip ≈ 158 credits. |
| Image | Nano Banana 2 | Generate or edit from text and images. About 8 credits per 1K image, 12 at 2K, 18 at 4K. |
| Image | Nano Banana Pro | Consistent run times across generations. About 12 credits per 1K or 2K image, 21 at 4K. |
| Image | Nano Banana 2 Lite | Faster, simpler variant. A flat 8 credits per image at every resolution. Start here to test. |
| Image | GPT Image 2 | The lowest-cost image model here. About 3 credits per 1K image, 6 at 2K, 12 at 4K. |
| Image | Seedream 5.0 Lite | Flat pricing — about 8 credits per image whether you generate or edit. |
Pick a year and the props come with it — the backdrop, the hair, the film stock. If 1996 lands wrong, 1988 is one word away.