One photo in · A pencil drawing out · Free in the browser
Graphite shading, cross-hatching and a loose paper-edge fade. Your photo comes back redrawn as a hand-drawn pencil portrait in seconds.
A real pencil rendering, drawn by the model behind this page. Drag the divider to compare the plain studio photo with the finished pencil sketch. Every output is AI-generated, so your result varies with your photo, your prompt, and the style you describe — treat this as a preview of what the tool can do, not a fixed result.
Drag to compare before and after
The tonal shading, the cross-hatching, the confident contour lines, the loose edges that dissolve into the paper — the kind of graphite sketch an artist would spend an afternoon on, redrawn from one ordinary photo.
Instead of commissioning a portrait artist or wrestling with a filter that just posterises the pixels, you upload a clear picture, name the drawing style you want, and the model rebuilds the image stroke by stroke while keeping the face and expression recognisable. The point of starting from your own photo is that the result still reads as the person, not a generic face wearing their haircut.
On Kavel this runs on the Nano Banana image model, so a finished drawing comes back in one generation rather than a week of waiting on a commission. You start from your uploaded picture — 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 style costs nothing but time.
It suits someone making a keepsake gift, a couple wanting their engagement photo as line art, a creator who needs a hand-drawn avatar, or anyone who likes the look of graphite but cannot draw. A photo filter flattens tone; this rebuilds the image as marks on paper, which is why the shading follows the face instead of the pixels.
And because the same studio also changes hair, swaps outfits and handles other photo edits, one workflow covers a range of looks without a new tool for each.
Creative engine
Upload a photo, name the style, choose the format and resolution, and the credit estimate updates before you generate.
Upload one clear photo and the model reads the face, then redraws the whole frame as graphite marks on paper.
Name the style — soft tonal shading, clean line art, loose charcoal — and the drawing follows that brief.
Set the format, check the credit estimate, and generate; a failed run is refunded (unless it broke the content policy), so a second style is free to try. A sketch keeps the proportions honest. The caricature deliberately breaks them, and the pet portrait points the same drawn treatment at an animal.
From one plain photo to a finished graphite portrait, with the cost shown before you commit.
Add one clear picture with the face visible and evenly lit. Good contrast between light and shadow gives the model the tonal range it needs, and a sharp original keeps the contour lines confident rather than mushy.
Set the brief: soft tonal graphite, tight cross-hatching, clean single-weight line art, loose charcoal with smudged shadows. Specific direction keeps the drawing on-style instead of generic.
Confirm the estimate, generate the drawing, and download it to print or frame. If a run fails, it is free, so you can adjust the brief and try another style.
Gift prints, tattoo references, a portfolio that needs a hand-drawn page.
Turn a wedding photo, a pet, or a grandparent's portrait into a graphite piece worth framing — the kind of gift that lands because it was drawn from their own picture.
Get a hand-drawn profile image that stands out against a wall of photos, generated in minutes rather than commissioned and waited on.
Block out a tonal study from a reference photo to see where the shadows fall before committing graphite to real paper.
Common questions about turning a photo into a graphite portrait with Kavel.
No, and the difference shows in the shading. A filter runs an edge-detect over the pixels and leaves you with a grey, flattened photo, not a sketch. This rebuilds the image as marks — the tone follows the bone structure, the strokes follow the hair direction, and the edges fade out the way a real hand stops drawing. That is why it survives being printed, and a filtered photo usually does not.
A sharp picture with the face evenly lit and real contrast between light and shadow. Graphite is entirely tone, so a flat, evenly-lit snapshot gives the model nothing to shade with, and a dim or blurry original makes the contour lines uncertain. Front-facing portraits with a plain background work best.
Yes. Describe it and the artwork follows — soft tonal graphite, tight cross-hatching, clean single-weight line art, loose charcoal with smudged shadows, or a quick gesture study. A clearer, more specific brief gives a closer result than a vague one, so it helps to name the medium and the level of detail rather than leaving them to chance. If the first pass is close but not quite there, adjust one detail and generate again.
That is the whole reason for starting from a real photo — the model keeps the face, the hair and the expression while it redraws them in graphite. A clear, well-lit original holds the likeness best; a dark or low-resolution one gives it less to work from, so the closer the input, the closer the drawing.
It depends on the resolution you pick and how many versions you generate, and the estimate is on screen before you press the button. Free credits arrive with the account, and a failed run is refunded automatically (unless it broke the content policy) — so you can try tonal graphite, then line art, then charcoal, without watching a meter tick up on every attempt.
Yes, and it is worth generating at a higher resolution first if you plan to. Graphite detail lives in fine strokes, and a 1K file will look fine on a phone but will lose the cross-hatching once it is enlarged to frame size. Pick the largest resolution the estimate offers, download the file, and print it on a matte or textured stock — the paper grain does a lot of the work.
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. |
The model behind the drawing, and other ways to restyle a photo.
The drawing comes back in seconds, and if the graphite reads too soft or too hard, the next style is one word away.
A real edit made with this tool — drag your own photo in on the left, the prompt is already written, then press Generate.
