Turn Yourself into a GTA 6 Character with AI

GTA 6's two main characters are Jason and Lucia, a couple pulled into a conspiracy across the state of Leonida (Xbox), and the game launches on November 19, 2026 (Rockstar Newswire). If you'd rather see yourself as a GTA 6 character, an AI image model can redraw your photo in the game's painted style.
This guide shows the prompt we tested with GPT Image 2.6, the before/after result, six character ideas to try, and how to turn the portrait into a moving shot on DramaCanvas. Reading takes about 7 minutes.
How do you turn your photo into GTA style, step by step?
Connect your photo to a GPT Image 2.6 node, paste the prompt from this article and generate:
Open a canvas on DramaCanvas and drag your photo onto it.
Click Add node → Image, choose GPT Image 2.6, and drag a line from your photo into the new node.
Paste the prompt below, type @ and pick Image 1, then generate. New accounts get free credits to try it.
Which photo works best?
The best photo is a clear, well-lit shot of one person facing the camera or turned slightly, with shoulders visible and eyes uncovered.
- Light: even daylight on the face. Hard shadows can turn into stubble or marks in the drawing.
- Eyes: no sunglasses or low caps. If you want sunglasses, ask for them in the prompt.
- Framing: head and shoulders at least, so the model has a pose to keep.
- Consent: only use photos of yourself or of people who agreed, and never of public figures.
What does the GTA 6 art style look like?
The GTA 6 style is modern comic key art: crisp black ink lines, semi-realistic faces, glossy cel-shaded painting and a hot pink and orange sunset palette, set in present-day Florida. It descends from the pop-art covers Stephen Bliss painted for the series from GTA III onward (IBTimes UK), and Rockstar's GTA VI box art keeps that comic-panel look (Creative Bloq).
This is the style block we put into every prompt:
modern stylised comic key-art illustration like a contemporary open-world crime game cover: crisp clean black ink linework, semi-realistic proportions and faces, smooth digital painting with cel-shaded gradient shadows and glossy highlights, hyper-saturated Florida palette of hot pink and orange sunset gradients, turquoise water and palm tree silhouettes, present-day 2020s setting (smartphones, modern sports cars, streetwear, social-media era)"Semi-realistic proportions and faces" matters most for a portrait: without it, faces drift toward anime or caricature, and you stop looking like yourself.
How do you write a photo-to-GTA prompt that keeps your face?
Write the prompt in four parts: a redraw instruction, the style block, a KEEP list and a CHANGE list, followed by a few negatives.
- Redraw: "Redraw this exact photo as a vertical game loading-screen portrait…"
- KEEP: face shape, hair, facial hair, smile, the clothes you're wearing, the pose.
- CHANGE: background, light and camera angle, and nothing else.
- Negatives: no text, no logos, not any existing game character.
The KEEP list does the heavy lifting. A style prompt alone pulls the face toward a "typical" game hero, while naming what to keep holds it close to your photo.
What did our before/after test show?
Our test kept the face, beard, curls, smile, shirt and chain, and turned the street background into a sunset beachfront. We used a street photo of a fictional man we generated for this test, not a real person, at 1024×1536. We generated our examples with GPT Image 2.6 through OpenAI's image API at high quality, so results on DramaCanvas can differ slightly.
Redraw this exact photo as a vertical game loading-screen portrait in a modern stylised comic key-art illustration like a contemporary open-world crime game cover: crisp clean black ink linework, semi-realistic proportions and faces, smooth digital painting with cel-shaded gradient shadows and glossy highlights, hyper-saturated Florida palette of hot pink and orange sunset gradients, turquoise water and palm tree silhouettes, present-day 2020s setting (smartphones, modern sports cars, streetwear, social-media era). Keep the same man: same face shape, beard, curly hair, smile, white linen shirt and thin gold chain, same pose. Change the background to a modern beachfront boulevard at sunset with palm trees, pastel art-deco hotels and a glowing glass-tower skyline; add a slight low camera angle. All characters are fictional, not any real person and not any existing game character. No weapons. No text, no letters, no numbers, no title, no logos, no brand marks, no readable signs or license plates, no watermark.- Kept: face shape, beard, curly hair, smile, white linen shirt, gold chain.
- Changed as asked: background, sunset light, low camera angle, ink-and-gloss finish.
- Not asked for: the model put a phone in his hand. If you don't want it, add "empty hands".
- No text or logos appeared.
How do you make it on DramaCanvas?
On DramaCanvas you do it in one Image node with your photo connected as a reference.
Step 1: Add your photo and an Image node
Drag your photo onto the canvas. Click Add node in the toolbar at the bottom and choose Image, then pick GPT Image 2.6 in the model picker under the node.
Step 2: Connect the photo
Drag a line from the connection point on the right of your photo node into the new Image node. In the prompt, type @ and pick Image 1 so the model knows which picture is you.
Step 3: Set the size and generate
Open Image settings, pick 9:16 or another portrait ratio under Size or aspect ratio, and start with Resolution at 1K (2 credits, about $0.05). When the likeness is right, run it again at 4K (4 credits, about $0.11). Failed generations are refunded automatically.
What GTA 6-style characters can you make?
Keep the KEEP list the same and swap only the CHANGE list to make a different character. We haven't tested these six, so treat them as starting points:
- Beach-club owner: open silk shirt and aviators, poolside cabanas at noon.
- Street-racer mechanic: grease-marked work shirt, a tuning garage lit in neon.
- Airboat tour guide: khaki vest, airboat in golden-hour sawgrass.
- Rooftop influencer: bright streetwear, phone on a gimbal, penthouse pool at dusk.
- Night-market chef: rolled sleeves and apron, food truck under string lights.
- Yacht-party DJ: white blazer and headphones, decks on a yacht at sunset.
How do you keep the same face across several images?
To keep the same face across several images, connect two references to every new Image node: your original photo and your first good result. In the prompt, say "face from Image 1, illustration style from Image 2" and write a new CHANGE list. More on this in keeping a character consistent.
Once you like your character, put them on a poster with the GTA 6 cover art guide, or on your phone with a GTA 6 wallpaper.
Can you make your character move?
Yes: connect the portrait to a Video node and describe the motion. Click Add node → Video and choose a model:
- Kling 4 uses your portrait as the Start frame (expand Parameters in the input box). It costs 4 credits per second at 720p, so 5 seconds is 20 credits (about $0.54).
- Seedance 2.5 uses it as a reference image and can stage a new scene with the same person. It costs 30 credits per second at 720p, so 5 seconds is 150 credits (about $4.05).
We tested Seedance 2.5 with an earlier portrait of the same test man (an 8-second shot at 720p and 16:9, 240 credits, about $6.48 at DramaCanvas prices):
Use image 1 only as the identity and illustration style reference. The entire environment has absolutely no writing, no signs, no shops, no billboards, no storefronts, no license plates. Only ocean, palm trees, sky and a clean empty road are visible in the background. A single cinematic game-trailer shot of the same adult man, with exactly the same facial features, hair and clothing, driving a fictional unbranded 1980s convertible along a palm-lined coastal boulevard at sunset. Both hands naturally control the steering wheel. Begin with the camera alongside the moving car for two seconds, then track forward and arc around the nose of the car, ending directly ahead of it looking back through the windshield at the man for the final two seconds, keeping his face readable. Rich violet and coral sunset, colored reflections, bold clean black outlines, painterly comic poster shading throughout. Stable anatomy, wheels and vehicle geometry, smooth continuous camera motion. No captions, text, logos, trademarks, watermarks or photorealism.In the frames we sampled, his face, curls, beard and shirt matched the reference, and the camera move followed the prompt. Our first run put shop signs with garbled lettering in the background. Saying "only ocean, palm trees, sky and a clean empty road" in the prompt fixed it, so describe a background with no signs when you want clean frames.
To cut several shots into one video, see how to make a GTA 6-style trailer.
Frequently asked questions
Can AI make me look like Jason or Lucia?
That's not what this guide does. Copying the game's characters isn't yours to do, and prompts that name them are more likely to be refused. The prompt here gives you an original character in the same art style, with your own face.
What happens to the photo I upload?
You upload it to your canvas and connect it as a reference. For how DramaCanvas handles your data, read the privacy policy.
Why was my prompt refused?
In its ChatGPT guide, Starrd reports that refusals come from asking for a trademark, like the logo, the Rockstar wordmark or a real car brand, while a style reference is fine (Starrd). If a prompt is refused, remove brand words and keep the style description.
Can I do a couple or a group?
Yes. GPT Image 2.6 on DramaCanvas accepts up to 8 reference images, so connect one clear photo per person and give each one a separate KEEP list. Everyone in the photos must agree.
DramaCanvas is not affiliated with Rockstar Games or Take-Two Interactive. To make your own character, open a canvas, drop in your photo, and connect it to a GPT Image 2.6 node with the prompt above.
Open a canvas with GPT Image 2.6 ready: new accounts get free credits.



