One photo in · A felt puppet out · Free to try, nothing to install
Fleece skin, yarn hair and big googly eyes, with your beard, glasses and expression still in there. That last part is what decides whether it lands.
A real felt-puppet render from the model this page uses. Drag the divider and check the resemblance: the ginger hair and beard become yarn and felt, the black glasses and burgundy sweater stay, the smile survives. Every output is AI-generated, so your result varies with your photo and the direction you give — treat this as a guide to what the tool can do, not a fixed result you will get back exactly.
Drag to compare before and after
Fuzzy fleece skin with visible fibre, an oversized rounded head, yarn hair, a wide hinged mouth, big googly eyes with painted pupils — one photo of you rebuilt as a felt hand puppet from a children's television show. The trick is not the puppet, it is the resemblance.
Anyone can generate a generic puppet; what makes the picture funny is that it is unmistakably you in fabric, so the model has to carry across the things people actually identify you by — your hair colour and how it sits, a beard, glasses, the clothes you had on, the particular way you smile.
The felt pass 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 — set the format and resolution, and the credit estimate updates before you commit; a failed run is refunded (unless it broke the content policy), so a second attempt with different direction is free.
The result works as a profile picture, a birthday card, a team page where everyone gets the same treatment, or a gift for someone who grew up on puppet shows. Distinctive features help enormously: glasses, a beard, unusual hair and a strong colour in your top all survive the translation into felt and give people something to latch onto.
A plain, evenly lit photo facing the camera converts best, and because the same studio also paints you as an oil portrait, draws you as a caricature and turns you into a collectible figurine, one workflow covers a whole shelf of styles. This makes a puppet version of your own face — it is not a way to generate copyrighted characters from a specific show.
Creative engine
Upload a photo, steer the puppet style, choose the format and resolution, and the credit estimate updates before you generate.
Rebuilds your face in fleece and yarn with real fabric texture rather than a flat cartoon illustration.
Carries across the features people recognise you by — hair, beard, glasses, clothing, expression.
From one photo to a fuzzy felt version of yourself, with the cost shown before you commit.
Face the camera in even light, close enough that your features read clearly. Glasses, a beard, distinctive hair and a strong sweater colour all survive the translation and make the result recognisable.
Say what you want — bigger googly eyes, a fuzzier fleece, a specific yarn hair colour, a full-body puppet rather than head and shoulders.
Generate, then download the puppet and see who recognises you. A failed run is refunded (unless it broke the content policy), so adjusting the direction and trying again costs nothing.
Streamer avatars, birthday jokes, a school project about puppets.
Swap your avatar for a fuzzy version of yourself that still reads as you at thumbnail size.
Give a birthday card a punchline, or run the whole team through the same treatment so a staff page has one consistent gag.
Anyone who grew up on felt characters on television recognises the look instantly — turning them into one lands every time.
Common questions about turning a photo into a felt muppet with Kavel.
That is the whole point, and it is what separates a good result from a generic one. The model carries across hair colour and shape, facial hair, glasses, clothing and expression, then rebuilds the rest in fleece and yarn. Distinctive features translate best — a beard, strong glasses, unusual hair colour — while a plain, evenly lit face gives it the most to work from.
No, and that is deliberate. This turns your own face into a puppet in the general style of felt television puppetry — fleece, yarn, googly eyes. It does not reproduce named characters from a particular series, which are somebody else's copyrighted designs. Your own face in felt is yours to use.
It is rendered to look photographed: fabric fibre visible in the fleece, real studio lighting on the surface, stitching and seams where a puppet would have them. That is why it reads as an object someone could actually hold rather than a flat illustration of one.
A straight-on, evenly lit shot from the chest up, with your face unobstructed. Harsh side shadow, a steep angle, sunglasses, or a busy background all cost the model the detail it needs to make the resemblance land. A plain backdrop and a clear expression convert best.
Yes — describe it in the prompt. Ask for larger or smaller googly eyes, a particular yarn hair colour, a fuzzier or smoother fleece, a full-body puppet with arms rather than head and shoulders, or a specific background. A run you do not like takes seconds to redo.
A single clear subject converts most reliably. Groups can work, but every extra face is another likeness the model has to preserve, so the resemblance gets weaker the more people are in frame. For pets, the pet portrait tool is the better-tuned starting point.
The credit estimate appears before you generate and updates with 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 try the look and judge it before spending anything.
Generate at the highest resolution for printing on a card or poster. For commercial use, check the terms and make sure you own the photo you uploaded; a puppet version of your own face is straightforward, a puppet version of someone else's is not yours to sell.
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. |
Other ways to turn the same photo into a character, and the model behind the edit.
Fleece, yarn and googly eyes are the easy part. Look at whether your beard, your glasses and your expression made it through.
A real edit made with this tool — drag your own photo in on the left, the prompt is already written, then press Generate.
