Describe the person · Get the whole figure · No sign-up to try
Ask most image models for a person and you get a portrait: cropped mid-thigh, feet missing, the outfit you described only half visible. This page is set up for the opposite — the whole figure in frame, from hair to shoes, with the pose, the clothes and the setting you actually asked for. Type the description, press generate, and the framing is handled for you.
Both images below were generated on Kavel while building this page, specifically to check that the framing survives a change of pose and background. Every output is AI-generated, so yours will differ in face, wardrobe and light; read these as the framing you can rely on, not as pictures you will get back.
A plain grey backdrop, a straight-on standing pose, a knit sweater and tailored trousers. The whole figure sits inside the frame with clear space above the head and floor visible under the shoes, and the proportions read as a real person rather than a stretched one — which is the usual failure mode when a model is pushed to fit a full body into a portrait-shaped canvas.
Open the image studio
The harder test: a walking pose on a wet pavement with a city behind it. The stride is caught mid-step, the long coat hangs correctly, and both feet are still in the picture. Backgrounds and motion are where full-figure framing usually breaks down, because the model has more to spend the canvas on.
Try your own descriptionA full body AI generator is an image generator pointed at one specific problem: keeping the entire person inside the picture. It matters more than it sounds. Image models are trained on a world of photographs where portraits outnumber whole-figure shots many times over, so "a woman in a camel coat" reliably returns a chest-up crop — and the coat you wanted to see ends at the frame.
Getting the whole body takes explicit instruction about framing, headroom, and where the feet sit, which is what this page supplies before your description ever reaches the model.
What comes back is a photograph-style image of a person who does not exist, framed head to toe, at whatever pose and wardrobe you specified. If you want to restyle a real photo instead of inventing someone, pose changer and outfit generator work on a picture you already have.
Creative engine
Describe the person, the pose and the setting. The framing is prewritten, the canvas defaults tall, and the credit estimate updates before you generate.
A standing person is taller than they are wide, so the canvas defaults to 3:4 rather than square. Asking for a whole body inside a square is what produces the shrunken, distant look.
At full-figure scale hands and feet occupy few pixels, so detail there is the thinnest part of the image. Two or three takes and keep the cleanest — a failed run is refunded automatically (unless it broke the content policy).
Three things to state outright. Leave any of them to the model and it will choose the safest, most photographed option — which is a portrait crop.
"Head to shoes inside the frame, space above the head, floor visible below the feet" is an instruction the model can follow. "Full body" alone is a label it has seen attached to plenty of half-length pictures.
Standing straight on, walking three-quarters, seated on a stool — a stated pose keeps the limbs where you expect them. Words like "confident" or "casual" describe a feeling and leave the body unassigned.
A standing person is taller than they are wide, so a 3:4 or 2:3 frame gives the figure somewhere to go. Asking for a whole body inside a square is what produces the shrunken, distant look.
Three jobs that a chest-up crop simply cannot do.
Clothes have to be seen on a body from collar to hem. A portrait crop hides the trousers, the shoes and the way a coat actually falls, which is most of what a customer is judging.
Height, build, posture and wardrobe are how a character reads on the page. Seeing them standing whole makes a description concrete in a way a face never does.
Posters, storefronts and app onboarding screens are built around standing figures with space around them. Generate the figure with room to spare and the crop stays your decision — the same reason a movie poster starts from a full-length shot.
What holds up, what still slips, and how to steer it.
Because most photographs of people are portraits, and a model reproduces the distribution it learned. Unless framing is specified, the likeliest picture of "a person" is chest-up. Stating the frame — head to shoes, headroom above, floor below — is what shifts it, and it is why this page writes that in for you.
They are much better than they were and still the weakest part of the image. At full-figure scale hands and feet occupy few pixels, so detail there is thin; generating two or three takes and keeping the cleanest is faster than fighting one image with extra prompt words.
Not on this page, which generates from text. Editing an existing picture is a different flow — pick image-to-image in the studio and start from your upload, or use one of the editing tools where the photo is the input rather than the description.
3:4 or 2:3 for a standing figure, which is what the two demo images use. A square crowds a standing person and pushes the model to shrink them into the middle; a wide frame leaves empty sides unless the setting is doing real work.
Not exactly. Each run samples fresh, so the face changes even with identical wording. Keeping wardrobe, hair and setting fixed in the prompt gets you images that belong together, but a consistent identity across many images is a different job than this page does.
The free engine costs nothing to start and needs no account. Larger models spend credits, and the estimate for the model and resolution you picked appears in the generator before you start; a failed run is refunded automatically (unless it broke the content policy).
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. |
This page invents a person from text. These three change one you upload.
One line about who they are, what they are wearing and how they are standing. The framing is already set up.
A real edit made with this tool — drag your own photo in on the left, the prompt is already written, then press Generate.
