A version for a man
Replace "the woman in this photo" with "the man in this photo" in both prompts, change the pronouns, describe your hair and beard instead of the long hair and freckles, and swap the tank top for your own outfit. Not tested.
Your selfie redrawn like GTA cover art: a beach scene at sunset and a night scene under neon lights. 38 credits for both, demo flaws shown.
Real workflow: a selfie of a woman with long brown hair in a white T-shirt against a beige wall, then the same woman drawn in front of a turquoise convertible at sunset, and sitting on a black motorbike in a neon-lit street. On the beach, the car has its windscreen on one side and its bonnet on the other. At night, the picture is framed full length and the face is too small to recognise well.
At a glance
A GTA style photo is a portrait redrawn like the artwork on the series' covers and loading screens: black ink outlines, flat saturated colour, hard shadows. On 20 November 2026, the "Trending Now" list on viralvideoai.app had five GTA formats out of nineteen, including "GTA 6 Girlfriend Lucia AI" and "GTA Wasted Meme AI", with no date on the cards. The same day, videosi.io/trending listed "GTA 5 Goodbye AI" with the status "Spreading", since 11 November 2026.
Those formats are videos that put your face into a scene from the game. This template does something else: a still image, with no scene and no character from the game. Upload a selfie, get two portrait illustrations back. The first puts you in front of a convertible by the sea at sunset. The second sits you on a motorbike at night in a street of neon signs. Both demo renders were kept on the first try, flaws included: a flatter line than the cover art, a car whose perspective does not add up, and at night a face too small to recognise well.
For the video version, where you are inside the game, see put yourself in a video game. For a drawing with no game setting, there is photo to drawing. Both renders cost 38 credits, see pricing.
These are the files of the workflow above, exactly as the models produced them, unretouched. The starting point is shown when there is one.
Starting point
Render
People who want to see themselves drawn in the style of GTA cover art, for a profile picture, a wallpaper or a gift for a friend, without knowing how to draw.
A sharp selfie of you or of someone who agrees, front-facing and taken in daylight, plus the right person, the features to keep, the outfit and the vehicle in the prompts. The photo is optional.
Upload a sharp, front-facing selfie
Taken in daylight, with nothing across your face. The model simplifies your features to draw them, so the clearer the photo, the more it has left to recognise you by. The demo starts from a head-and-shoulders selfie against a beige wall.
Describe yourself in the prompt
Both prompts say "the woman in this photo" and mention her freckles and "long straight dark brown hair with a centre parting", which describe the demo. Put in the right person, your hair and whatever makes you recognisable, and change the pronouns if needed.
Change the outfit and the vehicle
The palm-print shirt, the turquoise convertible, the leather jacket and the black motorbike are all spelled out. Swap them for what you want to wear and drive, and keep it a generic model with no badge and no number plate. The sentence about the drawing style and the list of things to leave out can stay.
Compare at full size, then rerun one render
Zoom in and check the face, the hands and the vehicle. That is where the demo went wrong: a car facing two ways, an extra boot under the bike. If one render does not work, rerun only that one for 19 credits.
The same canvas, tuned for different needs: each variation is two or three lines of prompt away.
Replace "the woman in this photo" with "the man in this photo" in both prompts, change the pronouns, describe your hair and beard instead of the long hair and freckles, and swap the tank top for your own outfit. Not tested.
Replace "pastel turquoise 1980s convertible" or "black sport motorbike" with a description of your vehicle: type, colour, era. Keep "invented generic model with no badge and no number plate", since the prompt should not name a make or a model. Not tested.
Replace "framed from the hips up" with "framed from the shoulders up", take out the pose and the vehicle, keep the setting behind you, then switch the node from 3:4 to 1:1. The face then fills the picture, which is what the demo's night scene lacks. Not tested.
Replace the seaside boulevard sentence with your own scene, for example "a desert gas station at noon" or "a rooftop above the city at dusk". Change the light and the sky colours too, or the model will keep the pink and orange of the sunset. Not tested.
Upload a photo of the two of you, write "the two people in this photo" and describe each in one sentence: hair, outfit, position in the frame. Both people must agree to it, and each face will be smaller, so less faithful. Not tested.
Switch the node from 3:4 to 9:16 before you run it and ask for the top third to be left to the sky and the palm trees, where the phone shows the clock. On a taller frame the model may go full length, as it did in the night scene. Not tested.
The long dark brown hair, the centre parting and the freckles belong to the example portrait. With your photo, the model gets two descriptions that contradict each other and may draw you with her hair. Rewrite the "Keep her recognisable" sentence with your own features.
The night prompt asked for a frame from the knees up and the model drew the whole motorbike, full length. The face is then down to a few lines. If the likeness matters more than the vehicle, ask for a frame from the shoulders or the waist up.
The prompt asks for hard-edged painted shadows in two or three tones. Both demo renders have a clean, fairly flat vector line, with a white edge around the figure in the beach scene. You can tell which style it is going for, but it does not have the paint texture of the series' artwork.
The template does neither: the prompts rule out the title, the logo, the interface and existing characters, and those lines are best left in. On the demo, even neon signs requested without letters came out with fake ones. If you want a name on the picture, add it afterwards in an image editor.
In a closer shot, mostly yes. In the demo's beach scene the face shape, brown eyes, freckles and centre parting are kept, with thicker eyebrows and simplified features. In the night scene, framed full length, the face is small and you recognise her mainly by her hair.
The model drew the windscreen and the cabin on the woman's left and the bonnet and headlights on her right, as if the car faced two directions. The prompt asked for a parked convertible with her leaning on the door. We kept the picture as it came out so you can see it.
No. The prompts rule out the title, the logo, the minimap and any interface, as well as existing characters, and none of them appear on the demo. One caveat: at night, some of the neon signs look like fake letters, although the prompt asked for abstract shapes.
38 credits, 19 per render on Nano Banana Pro at 2K. Both demo renders were kept on the first try. If one of yours does not work, rerun just that one for 19 credits.
Yes, by switching the node to Nano Banana 2.1 at 1K, which costs less per render and gives smaller pictures. We did not test it on this template, and the likeness of the face is the part most likely to suffer. The demo runs on Nano Banana Pro.
1792 x 2400 pixels, portrait, in a 3:4 ratio. Enough for a profile picture, a phone wallpaper or a print of about 15 x 20 cm.
No, two still images. The GTA formats listed by the sites we cite are videos that put your face into a scene from the game. Here you get an illustrated portrait. You can plug a Video node in after the image in the canvas. We have not tested that on this template.
We do not know who started it. On 20 November 2026, the "Trending Now" list on viralvideoai.app had five GTA formats out of nineteen, with no date on the cards, and videosi.io/trending listed "GTA 5 Goodbye AI", added that day, status "Spreading", since 11 November 2026. Those are our only two sources, and Imaginode has no link with either site.
The photo is optional in the canvas, but without it the model invents a face. For the drawn character to be you, it needs your selfie.
No. "GTA" is used here to name a drawing style people search for under that name. Imaginode has no link with Rockstar Games, and the renders contain no logo, title, interface or character from the game. The template is meant for personal use: a profile picture, a wallpaper, a gift for a friend, not a commercial poster in the game's name.
GTA style AI photo: your selfie as a cover art illustration
Use this workflowFree account, no card required. The workflow opens pre-filled in your canvas.