One photo in · Your own head shape · Nothing to download
See what a cleanly shaved head looks like on you, with a natural head shape, your own skin tone across the scalp and everything else left where it was. Runs online, free to start, nothing to install.
A real shave, rendered by the model this page uses. Drag the divider and watch what holds still: the same face, the same beard, the same grey t-shirt, the same light. Only the hair goes. Every output is AI-generated, so your own result varies with the photo you upload and how much of your hairline it shows — read this as a guide to what the tool does, not a fixed result you will get back.


Drag to compare before and after
It takes one photo of you and removes the hair from your head, rebuilding the scalp underneath so it reads as a real shaved head rather than a smudge where the hair used to be. Three kinds of people search for it. One is thinking about shaving and wants to see the result before the clippers come out. One is losing hair anyway and wants to know whether going all the way looks better than holding on.
One just wants the picture for the group chat. All three want the same output: their own face, no hair, right now. The shave runs on the Nano Banana image model, in the browser, with free credits to start.
Upload a clear photo — the flow defaults to image-to-image — pick the format, and the credit estimate updates before you commit; a run that fails is refunded automatically (unless it broke the content policy). What decides the quality is the hairline. A photo shot straight on in even light, with your forehead and the front of your head unobstructed, gives the model the information it needs to guess the skull shape correctly.
A hat, a heavy fringe, or a steep angle from below hides exactly the part it has to reconstruct, and the result drifts. Facial hair is worth keeping in the frame on purpose: a beard is the strongest cue that the person in the after picture is still you.
Creative engine
Upload a photo, say how short you want to go, pick the format, and the credit estimate updates before you generate.
Rebuilds the scalp from your own head shape and skin tone, with a natural shadow where the hairline used to sit.
Works in the browser with free credits — no camera app, no account wall before you see a result.
Also runs shorter than shaved: ask for a buzz cut, a receding hairline, or a fully clean scalp. If the bald filter answers no, the hairstyle changer tries a different cut on the same photo, and the hair growth clip runs the bald filter backwards as video.
From one photo to a believable preview, with the cost shown before you commit.
Face the camera in even light with your forehead visible and no hat. The front of your head is the part being rebuilt, so it needs to be in view.
Ask for a fully shaved scalp, a number-one buzz, or a shaved crown with the sides kept — the wording in the prompt decides how far it goes.
Generate, then decide whether the clippers are actually coming out. A failed run is refunded (unless it broke the content policy), so trying a second length costs nothing but the wait.
People deciding, people already losing it, people just messing about.
Hair grows back slowly. Seeing your own head first is cheaper than finding out in the mirror three seconds after the first pass.
If the hairline is going anyway, compare the thinning version with the clean version on your own face rather than on a stranger in a stock photo.
Run a friend's photo, send it, deny everything. No app install, no watermark stamped across the picture.
Common questions about the bald filter and what it does to a photo on Kavel.
Yes, and long hair is often easier than a short crop, because the outline of your skull is clearer to the model when there is a definite mass of hair to remove. What matters more is whether your forehead and hairline are visible — hair swept forward over the brow hides the exact edge the scalp has to be rebuilt from.
Yes, unless you ask for it to go. Facial hair is left alone by default, which matters more than it sounds: a beard is the strongest signal that the person in the result is still recognisably you, so keeping it in the frame makes the preview far more convincing.
Yes — say which one you want. A number-one or number-two buzz, stubble on the crown, a shaved head with the sides kept, or a fully clean scalp are all different prompts, and each one costs the same to run. Trying two lengths and comparing them is the usual way people use this.
That is the whole question this answers, and the useful part is that it answers it on your head rather than someone else's. What decides the result is not the hair coming off but what is underneath: the shape of the skull, how far the hairline has already moved, and whether the scalp is a different tone to the face. The preview renders all three, which is why people are sometimes surprised — a shaved head reads as a stronger jaw on some faces and a longer head on others, and you cannot predict which from a mirror with hair in the way.
It is an estimate, not a scan. The model infers the skull underneath from the outline of your hair, your ears and your brow, which works well from a straight-on photo in flat light and less well from a low angle or a heavy side profile. It is close enough to answer the question you came with — it is not a measurement.
Technically yes, and this is where it is worth pausing. Editing a picture of a friend and sending it around is normal; publishing an altered photo of someone who has not agreed to it is a different thing, particularly on a subject as personal as hair loss. Use the same judgement you would with any photo of another person.
Because the model has to invent skin it has never seen. Shaved scalps carry faint stubble shadow, pores and a slight sheen; when the source photo is soft, over-compressed, or heavily beautified, there is little texture to carry over and the result can read as plastic. Asking directly for visible stubble shadow and natural skin texture usually fixes it in one retry.
The estimate follows the format and resolution you choose, so you see the cost first. Failed runs are refunded automatically (unless it broke the content policy), and new accounts start with free credits — enough to see yourself without hair before spending anything.
Your upload is used to produce your image, and the privacy policy has the specifics on retention. Worth reading before you run someone else's picture: a hair-loss preview of a friend is exactly the kind of image people would rather decide about themselves.
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. |
Say how short to go and the question stops being hypothetical. Nothing is committed, and nothing grows back slower for having looked.