1
Step 1
Photograph the hairline
Face the camera in even light with your forehead uncovered. Push your hair back if it falls forward. The temples and the crown are what the preview is built from, so they need to be in the frame.
One photo in · Your own head shape · Nothing to install
The clippers do not have an undo. Upload one photo and see your own head shaved down to a grade 1 first — same face, same ears, same jaw, same light. The only thing that changes is the hair.
Made with the model this page uses. Drag the divider — the jaw, the ears and the light on the left cheek are identical on both sides; only the hair is gone.
Drag to compare before and after
It takes one photo and cuts the hair down to clipper length, rebuilding the scalp underneath with your own skin tone and the same light that is already on your face. Everything else — features, expression, beard, clothing, background — is supposed to survive untouched. When it works, the picture is boring in the best way: it just looks like a photo of you with short hair.
The question underneath this is almost never fashion. It is people whose hair is thinning and who are wondering whether to stop fighting it — the ones who usually try the bald filter next, people about to enlist or start chemo, and people who have simply had the same head of hair for fifteen years.
What all of them want is not a haircut idea — it is one look at the version of themselves on the other side of a decision that takes ten minutes to make and three months to undo.
Different grades are genuinely different haircuts, and this is where one run is not enough. A grade 0 shows scalp and reads as a deliberate shave. A grade 2 keeps enough density to look like a choice rather than a surrender. A grade 4 is short back and sides territory and barely registers as buzzed at all.
Most people arrive wanting one number and leave having run three, because the difference between them on their own head is bigger than they expected.
Beards change the maths. Take the hair off a bare face and the head reads longer; take it off with a full beard and the whole silhouette rebalances. If you have facial hair, run the preview with it left in — the prompt keeps beards by default — because that is the combination you would actually be walking around with.
The photo matters more than the settings. A straight-on shot in even daylight with your forehead visible gives the model the hairline, the temples and the crown, which is everything it needs. A photo in a cap, in hard overhead sun, or from below gives it almost nothing, and the result will look like a good render of somebody else.
Creative engine
Upload a photo, say which grade you are thinking about, and the credit estimate updates before you press anything. The run happens on this page and the result opens directly under the box.
Hair hides the back and sides of a head. Cut it in a photo and the model has to invent what was under there — and if it invents a rounder or flatter skull than yours, the result is a stranger and the preview is worthless. This is the one thing worth checking in the output before anything else.
A receding hairline does not disappear at grade 1, it becomes more visible. A preview that quietly gives you back the hairline you had at twenty is flattering and useless. The honest version keeps the temples where they actually are, which is exactly what the people asking this question need to see.
From one photo to a version of your own head you can look at properly.
Face the camera in even light with your forehead uncovered. Push your hair back if it falls forward. The temples and the crown are what the preview is built from, so they need to be in the frame.
The prompt is already written for a grade 1. Change the number if you are considering a different length, and add anything the model should keep — a beard, glasses, a scar you do not want smoothed away.
Look at the shape of the head before you look at the haircut. If it is yours, run the grade either side and compare all three. That comparison is usually what settles it.
The people who ask this question are mostly not asking about fashion.
Deciding whether to get ahead of a receding hairline instead of managing it. Seeing the honest version — temples where they really are — is the point, not a flattering one.
Basic training, a role, a treatment starting next month. Looking first turns something that is going to happen anyway into something you have already seen.
No crisis, just curiosity that has never been worth the risk of acting on. Ten minutes with a photo answers it without a three-month grow-out.
What it keeps, what it needs, and what it will not flatter.
Upload one straight-on photo on this page, pick the clipper grade you are considering, and press generate. The result appears under the box on this page. There is nothing to install and you do not need to hand your photo to an app store first.
That is the thing to check first. Hair hides the back and sides, so the model rebuilds them from your proportions — the width of your jaw, the set of your ears, the slope of your forehead. If the head in the result does not look like yours, the photo probably did not show enough of your hairline. Retake it straight on with your forehead clear and run it again.
No, and it should not. A preview that quietly restores the hairline you had at twenty is pleasant to look at and no use for the decision you are making. The temples stay where the photo says they are.
Grade 0 leaves stubble and shows scalp. Grade 1 is the classic buzz — visibly shaved but with even coverage. Grade 2 keeps enough density to read as a choice. Grade 4 is closer to a short back and sides. They look more different on a real head than the numbers suggest, which is why running two or three is normal.
Yes. The prompt only touches the hair on your head, so a beard, stubble or moustache stays exactly as photographed. That matters, because the balance between a shaved head and facial hair is most of what the final look actually is.
The run happens when you press generate and the result is yours to download. If you are signed in it stays in your history so you can compare grades later; signed out, it is gone when you leave the page. Either way nothing is posted anywhere.
Technically it will run on any portrait. Whether you should is between you and the person whose head you are shaving — most people send it to them rather than about them.
Almost always the input. A hat, hair falling across the forehead, hard overhead light, or a low camera angle leaves the model guessing at the parts of your skull it cannot see. Even daylight and a straight-on angle fix it more reliably than any change to the prompt.
Yes — take the result into a [video model](/video/seedance-2) and it will animate, so you can see the cut from more than one angle before you commit. That is a separate run with its own estimate.
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. |
More ways to change the hair in a photo, and the model doing the work.
Ten minutes with a photo against three months of growing it back out. The estimate is on the button before you press it.
Examples made on Kavel with this model.
