GTA V to GTA VI, made with AI
A fan-made goodbye to GTA V with the full step-by-step workflow.
Complete production workflow • character references • scene prompts • Seedance motion prompts • edit timeline
PROJECT IDEA
A photorealistic live-action farewell to Michael, Franklin and Trevor, built as fast nostalgic memories that slow into an emotional Vinewood goodbye, then hand off into a GTA VI-style new beginning.
Built for 9:16 • Nano Banana Pro → Seedance 2.5 → fast-cut edit
00 At a glance
The exact workflow we settled on, stripped down to the decisions that mattered.
# · Step · What we did
1 · Lock the visual language · Photorealistic live-action GTA. Use GTA screenshots/art only as identity, environment and composition references, not as rendering style.
2 · Build clean character references · Ayham + Michael + Trevor + Franklin first. Portrait identity is more valuable than a complicated turnaround sheet. Add full-body only when useful.
3 · Collect iconic environment / prop references · Trevor’s truck, his trailer, Michael’s mansion/tennis court, heist planning room, Vinewood sign, GTA VI reference art.
4 · Generate FINAL stills before video · Every scene should already look like a finished movie frame. Do not ask Seedance to invent the scene from scratch.
5 · Animate each still for 8 seconds · Use one main action + one secondary reaction + camera movement. Cut the best 2-3 seconds. Final dialogue scene gets 8-10 seconds.
6 · Audio rule · Most GTA V memories use no dialogue: just laughter, engines, room tone, city ambience, wind, etc. Planning scene may have Michael speaking. GTA VI ending has the key dialogue.
7 · Edit order · Chaos → friendship → crew → heist → celebration → peace → goodbye → new beginning.
FORMAT SETTINGS: Portrait (9:16). Generate the stills vertically. Animate in Seedance 2.5 at 9:16. We chose 8-second generations so there is enough movement to extract the strongest 2-3 seconds without giving the model too much time to drift.
01 Character reference system
The final approach: simple, face-first, reference-driven.
WHY WE CHANGED APPROACH: We initially explored multi-angle character sheets. We then simplified: a clean portrait is the strongest identity anchor, because face consistency matters more than a complex sheet for these short clips. For the GTA characters, uploaded GTA images became the source of truth.
Michael De Santa: realistic portrait prompt
Using the uploaded image of Michael De Santa as the primary visual identity reference, recreate the exact same character as a highly photorealistic live-action human. Preserve his recognizable facial proportions, head shape, hairstyle, receding hairline, eyebrows, eyes, nose, lips, jawline, age, skin tone, and overall expression from the reference image. Do not redesign, beautify, modernize, or significantly alter his face. He should instantly read as the same Michael De Santa from Grand Theft Auto V, simply translated from a video-game character into a believable real person. Create a close-up head-and-shoulders portrait, facing mostly toward the camera with a calm, mildly irritated, slightly tired expression. Use realistic middle-aged skin texture, natural pores, subtle wrinkles, individual hair strands, realistic eyes, and believable human anatomy. Neutral dark-gray studio background, soft cinematic portrait lighting, sharp facial detail, natural lens perspective with no distortion. Visual style: premium photorealistic live-action GTA V movie adaptation, realistic enough to look like a photograph of a real actor. Not illustration, not cartoon, not cel-shaded, not game graphics, not CGI-looking. Facial likeness to the uploaded reference is the absolute highest priority. No text, no watermark, no props.
Trevor Philips: realistic portrait prompt
Using the uploaded image of Trevor Philips as the primary visual identity reference, recreate the exact same character as a highly photorealistic live-action human. Preserve his recognizable facial proportions, heavily receding hairline, thinning messy hair, forehead shape, intense eyes, nose, jawline, rough stubble, weathered skin, age, and distinctive unstable expression exactly as shown in the reference. Do not beautify him, clean him up, make him younger, or redesign his appearance. He should instantly read as Trevor Philips from Grand Theft Auto V, translated into a believable real-world person. Create a close-up head-and-shoulders portrait, facing mostly toward camera with an intense, unpredictable expression and the faintest unsettling half-smirk. Show realistic weathered skin, natural wrinkles, pores, individual stubble hairs, thinning hair, and sharp eyes. Neutral dark-gray studio background, soft cinematic portrait lighting, natural lens perspective, extremely clear facial detail. Visual style: gritty premium photorealistic live-action GTA V movie adaptation, realistic enough to look like a photograph of the real Trevor Philips. Not illustration, not cartoon, not cel-shaded, not game graphics, not exaggerated CGI. Facial likeness to the uploaded reference is the absolute highest priority. No text, no watermark, no props.
Franklin Clinton: realistic portrait prompt
Using the uploaded image of Franklin Clinton as the primary visual identity reference, recreate the exact same character as a highly photorealistic live-action human. Preserve his recognizable facial proportions, short cropped hairstyle, forehead, eyebrows, eyes, nose, lips, jawline, skin tone, age, and overall confident expression from the reference image. Do not redesign his face, change his age, hairstyle, or facial structure. He should instantly read as the same Franklin Clinton from Grand Theft Auto V, simply translated into a believable real person. Create a close-up head-and-shoulders portrait, facing mostly toward camera with a calm, confident, slightly serious expression. Use extremely realistic skin texture, natural pores, subtle facial detail, individual hair strands, sharp realistic eyes, and believable human anatomy. Neutral dark-gray studio background, soft cinematic portrait lighting, natural portrait-lens perspective with no distortion. Visual style: premium photorealistic live-action GTA V movie adaptation, realistic enough to look like a photograph of a real actor portraying Franklin. Not illustration, not cartoon, not cel-shaded, not gameplay graphics, not CGI-looking. Facial likeness to the uploaded reference is the absolute highest priority. No text, no watermark, no props.
GTA VI: Jason + Lucia
SOURCE OF TRUTH RULE: For the final GTA VI scene, the uploaded Jason/Lucia references override the written styling. If Jason is shirtless with sunglasses, keep him shirtless with those sunglasses. Do not invent new clothes. Copy visible hair, accessories, tattoos, jewelry, body type and styling from the references.
02 Environment & prop references
The trick that made the scenes read instantly as GTA: give the model visual anchors.
Reference · How we used it
Trevor’s truck · We stopped trying to generate it from text alone. We found a clear internet reference of Trevor’s actual truck and used that directly as the vehicle reference.
Trevor’s trailer / Sandy Shores · Use a GTA screenshot as environment reference. Character refs still control identity.
Michael’s mansion / tennis court · A GTA screenshot helps lock the architecture and court layout. Amanda + male trainer can stay secondary in the background.
Heist planning room · Use the GTA heist-planning screenshot for room layout, board, maps and atmosphere.
Vinewood sign · Use the screenshot as the exact landmark/location reference for the goodbye.
GTA VI art · Use the uploaded art/reference for Jason/Lucia identity and Vice City energy. Translate it into real-world materials and human skin, not illustrated rendering.
GENERATION RULE (applies to every Seedance step below): Generate ~8 seconds (10 seconds for the GTA VI dialogue ending), then cut the best 2-3 seconds. Use one clear action, real body motion and camera movement. Preserve the still; do not ask Seedance to rebuild the scene.
SCENE 01: Trevor: Sandy Shores Truck Chaos
Immediate energy. Trevor driving fast, both laughing, police lights far behind.
- References: Ayham reference + Trevor reference + Trevor’s actual truck reference
- Best cut: 2-2.5 sec: Trevor glances at Ayham → both laugh → truck bounces
- Audio: No dialogue. Engine, dirt-road rattles, laughter, distant siren optional.
A. Final still: Nano Banana Pro
Create a cinematic photorealistic live-action Grand Theft Auto V scene using the provided reference images of Ayham, Trevor Philips, and Trevor’s exact truck. Preserve Ayham’s exact facial identity, bald head, beard, skin tone, age, and body proportions from the reference, but change his clothing to suit the Sandy Shores setting. Dress Ayham in a faded dark charcoal T-shirt, dusty olive overshirt worn open, worn dark jeans, and rugged boots. Preserve Trevor Philips’ exact recognizable facial identity, receding hairline, messy hair, stubble, lean build, and signature worn clothing. Preserve the exact design, proportions, color, damage, and distinctive details of Trevor’s truck from the provided vehicle reference.
Scene: Trevor is driving his truck fast through Sandy Shores on a dusty desert road while Ayham sits in the front passenger seat beside him. The camera is positioned inside the truck near the center of the dashboard, creating a cinematic wide two-shot that clearly shows both faces and upper bodies.
Trevor is gripping the steering wheel with one hand and laughing hard with a wild grin. Ayham is laughing beside him, leaning slightly with the movement of the truck, as if they have just survived something completely stupid and chaotic. Their chemistry should feel like two friends sharing an insane GTA moment.
Through the truck windows, show a recognizable Sandy Shores atmosphere: dry desert terrain, dusty road, scattered trailers and low buildings, utility poles, distant mountains, and warm golden late-afternoon sunlight. In the far background, faint red-and-blue police lights are visible through the dust, suggesting a chase without dominating the scene.
The truck interior should look old, dirty, battered, and authentic to Trevor’s vehicle. Add dust in the air, warm sunlight entering through the windows, realistic reflections on the windshield, subtle road motion energy, and cinematic depth of field.
Visual style: photorealistic cinematic live-action adaptation of Grand Theft Auto V, gritty Southern California desert atmosphere, high-budget action-comedy movie cinematography, realistic skin, realistic vehicle materials, realistic clothing, warm natural sunlight, subtle GTA-inspired color grading. It should look like a real frame from a live-action GTA V movie, not a game screenshot, not illustrated promotional art, not cartoon.
Vertical 9:16 composition for TikTok, Instagram Reels, and YouTube Shorts. Keep Trevor and Ayham large and clearly recognizable in frame. No subtitles, no text, no logos, no HUD, no watermark, no extra people inside the truck, no duplicate characters, no distorted faces, no malformed hands, no incorrect truck design.
B. Motion: Seedance 2.5
Animate the provided image into an 8-second photorealistic live-action driving shot. Preserve Ayham, Trevor, their exact faces, clothing, truck interior, truck design, Sandy Shores environment, police lights in the distance, lighting, and composition.
Trevor drives quickly along the dusty desert road while laughing hard. The truck naturally bounces over the uneven terrain and Trevor maintains control of the steering wheel.
Ayham laughs beside him, his body moving naturally with the suspension and vibration of the truck.
Trevor briefly glances toward Ayham with a huge grin, then quickly looks back toward the road. Ayham looks toward Trevor and laughs harder.
Dust trails behind them. The distant police lights remain far behind and flash subtly through the dust. Trailers, utility poles, and desert scenery move realistically past the windows.
Add clear vehicle vibration, dashboard shake, realistic suspension movement, natural body bouncing, clothing motion, blinking, facial animation, dust movement, and mild camera vibration from the road.
No spoken dialogue and no lip-synced talking. Only natural laughter, engine sound, tire noise on dirt, suspension rattles, wind, and distant sirens if audio is generated.
No crash, no collision, no new passengers, no extreme stunt, no facial distortion, no truck redesign.
SCENE 02: Franklin: Drunk Night Outside Vanilla Unicorn
Friendship comedy. Franklin leans on Ayham while leaving the club.
- References: Ayham + Franklin identity refs. Club exterior comes from the generated scene.
- Best cut: 2-2.5 sec: Franklin stumbles/laughs → Ayham catches him
- Audio: No dialogue. Laughter, footsteps, nightlife ambience, traffic.
A. Final still: Nano Banana Pro
Create a cinematic photorealistic live-action Grand Theft Auto V scene using the provided reference images of Ayham and Franklin Clinton. Preserve Ayham’s exact facial identity, bald head, beard, skin tone, age, and body proportions from his reference. Preserve Franklin Clinton’s exact recognizable facial identity, haircut, skin tone, age, athletic build, and facial structure from his reference.
Scene: Ayham is helping a drunk Franklin Clinton walk out of a Los Santos strip club late at night. Franklin is visibly drunk, laughing hard, slightly off balance, and leaning heavily into Ayham for support. Franklin has one arm loosely around Ayham’s shoulders while Ayham holds him up and helps him walk. Franklin is carrying a bottle in one hand, holding it carelessly while laughing like he is having the time of his life.
Ayham is smiling and laughing at the situation while keeping Franklin from falling over. Their interaction should feel spontaneous, funny, and like two close friends leaving after an insane night out.
Dress Ayham in stylish Los Santos nightlife clothing: fitted black T-shirt, dark premium bomber jacket, dark jeans, clean black or white sneakers, and a simple watch. Dress Franklin in upscale casual streetwear that fits his GTA V personality: stylish dark jacket or overshirt, fitted shirt, dark jeans, fresh sneakers, and subtle jewelry. Franklin’s outfit should look slightly disheveled from partying.
The setting is directly outside a luxurious neon-lit Los Santos strip club at night. At the club entrance, include two glamorous adult women in their mid-to-late 20s, clearly adults, wearing fashionable nightlife outfits with very short fitted cocktail dresses and high heels. They are standing together beside the entrance, smiling and chatting while watching Ayham and the drunk Franklin stumble out. They should look glamorous, confident, stylish, and appropriate for an upscale Los Santos nightlife venue. Keep them as secondary background characters, clearly visible but not competing with Ayham and Franklin as the main subjects.
Behind them, show the club entrance with glowing pink, red, and purple neon lights, velvet ropes, a doorman or security guard, parked luxury cars, city streetlights, reflections across slightly wet pavement, and a lively late-night Los Santos atmosphere. The background should feel rich, chaotic, sexy, and unmistakably GTA nightlife without becoming overcrowded.
Camera composition: cinematic medium-wide vertical shot showing enough of both Ayham and Franklin’s bodies to clearly communicate Franklin leaning heavily into Ayham. Franklin’s drunken posture, bottle, and laughter must be immediately readable. Ayham and Franklin dominate the foreground. The two glamorous adult women and club entrance remain visible behind them and help establish the nightlife atmosphere.
Lighting: cinematic nighttime lighting with strong pink and purple neon glow from the club, warm streetlights, realistic reflections on skin and clothing, subtle dramatic shadows, and enough soft frontal illumination to keep Ayham and Franklin’s faces clearly recognizable.
Visual style: photorealistic cinematic live-action adaptation of Grand Theft Auto V, believable real human characters, gritty stylish Los Santos nightlife, high-budget crime-comedy movie cinematography, realistic skin, clothing, hair, neon lighting, natural anatomy, and subtle GTA-inspired color grading. It should look like a frame from a premium live-action GTA V movie, not gameplay footage, not cartoon, not illustration, not cel-shaded, and not obvious CGI.
Vertical 9:16 composition for TikTok, Instagram Reels, and YouTube Shorts. No subtitles, no text, no HUD, no logos, no watermark, no duplicate characters, no distorted faces, no malformed hands, no extra limbs, and no duplicated bottles.
B. Motion: Seedance 2.5
Animate the provided image into an 8-second photorealistic cinematic live-action shot. Preserve both characters’ exact facial identities, clothing, body proportions, nightclub exterior, background people, vehicles, lighting, and composition from the source image.
Franklin is slightly drunk and laughing hard while leaning heavily against Ayham for support. His arm stays naturally around Ayham’s shoulders as they slowly stumble forward together. Ayham laughs with him while keeping Franklin balanced. Franklin loosely carries the bottle in his lowered hand without drinking from it.
The two adult women near the entrance continue casually talking and occasionally glance toward them with amused smiles. The security guard remains naturally in position. Neon lights subtly flicker and reflect across the pavement and clothing.
Add visible natural motion: slow walking, shoulder movement from laughter, slight stumbling, head movement, blinking, breathing, clothing movement, and subtle handheld camera motion. The camera slowly tracks backward as Franklin and Ayham approach.
No spoken dialogue. No lip-synced talking. Only natural laughter, footsteps, distant nightlife ambience, subtle traffic sounds, and environmental sound effects if audio is generated.
Keep facial identity stable throughout. No falling, no new people, no exaggerated movements, no camera spins, no facial morphing, no extra limbs.
SCENE 03: Michael: Tennis at the Mansion
Comedy beat. Michael is irritated; Ayham is trying not to laugh.
- References: Ayham + Michael refs. Michael’s mansion/tennis court screenshot optional.
- Best cut: 2-2.5 sec: Michael side-eye → Ayham breaks into a laugh
- Audio: No dialogue. Small laugh, tennis-court ambience, birds.
A. Final still: Nano Banana Pro
Create a cinematic photorealistic live-action Grand Theft Auto V scene using the provided reference images of Ayham and Michael De Santa. Preserve Ayham’s exact facial identity, bald head, beard, skin tone, age, and body proportions from his reference. Preserve Michael De Santa’s exact recognizable facial identity, age, hairstyle, facial structure, and body proportions from his reference.
Change both characters’ clothing to suit a casual tennis session at Michael’s luxury Los Santos mansion. Dress Ayham in a fitted white tennis polo, black athletic tennis shorts, white ankle socks, clean white tennis shoes, and a simple sports watch. Dress Michael in a pale blue short-sleeve tennis polo, clean white tennis shorts, white socks, classic white tennis shoes, and a subtle luxury sports watch.
Scene: Ayham and Michael are standing beside each other on the private tennis court behind Michael’s Rockford Hills mansion, getting ready to play tennis. Both are holding tennis rackets naturally, with a few tennis balls nearby. Michael looks clearly irritated and mildly angry, giving Ayham a classic frustrated side-eye. Ayham is looking toward Michael and laughing, visibly amused by how annoyed Michael is.
In the background near the mansion and pool area, show Amanda doing yoga while a fit adult male trainer stands very close beside her, guiding her posture and stretching. The trainer must clearly be a man. Their interaction should feel slightly suspicious and awkward from Michael’s point of view, which explains why Michael is annoyed. Amanda and the male trainer should remain smaller in the frame and slightly out of focus, but still clearly visible and understandable.
The location should strongly evoke Michael De Santa’s iconic Los Santos mansion: luxurious Mediterranean-style villa architecture, palm trees, manicured landscaping, private swimming pool visible behind the tennis court, upscale hillside Los Angeles atmosphere, bright Southern California setting.
Camera composition: cinematic medium-wide shot from slightly off-center on the tennis court. Ayham and Michael dominate the foreground and both faces are clearly visible. Michael’s annoyed expression and Ayham’s laughter are the main focus. Amanda and the male trainer are visible over Michael’s shoulder in the background so the visual joke is immediately understandable.
Lighting: warm bright late-morning California sunlight, natural highlights on faces and clothing, soft realistic shadows, cinematic depth of field, realistic reflections, premium high-budget movie lighting.
Visual style: photorealistic cinematic live-action adaptation of Grand Theft Auto V, believable real human characters, upscale Los Santos atmosphere, subtle GTA-inspired color grading, realistic skin, hair, clothing, tennis equipment, architecture, landscaping, and natural sunlight. It should look like a frame from a high-budget live-action GTA V movie, not gameplay footage, not illustrated GTA artwork, not cartoon, not cel-shaded.
Vertical 9:16 composition for TikTok, Instagram Reels, and YouTube Shorts. Keep Ayham and Michael large and clearly recognizable in the foreground. No subtitles, no text, no HUD, no logos, no watermark, no duplicate people, no distorted faces, no malformed hands, no extra limbs, and no duplicated tennis rackets.
B. Motion: Seedance 2.5
Animate the provided image into an 8-second photorealistic cinematic live-action scene. Preserve Ayham, Michael, their exact facial identities, tennis outfits, rackets, mansion, court, tennis balls, and background people exactly as shown.
Michael looks toward Ayham with an increasingly irritated and suspicious expression. He shifts his weight slightly, lowers his racket a little, and gives Ayham a long classic annoyed side-eye.
Ayham notices Michael’s reaction, tries to keep a straight face for a moment, then breaks into a natural restrained laugh.
In the background, the adult woman continues a slow yoga movement while the adult male trainer calmly assists with positioning. Keep their movement subtle and secondary to Michael and Ayham.
Add natural blinking, breathing, small head movements, realistic hand movement on the rackets, a light breeze through clothing and palm leaves, and a very slow cinematic push-in toward Michael and Ayham.
No spoken dialogue. No lip-synced talking. Only a small natural laugh from Ayham, tennis-court ambience, birds, distant environmental noise, and subtle movement sounds.
No exaggerated reaction, no new people, no distorted rackets, no facial morphing, no camera spin.
SCENE 04: Trevor: Angry Outside the Trailer
Dark comedy. Trevor is furious; Ayham is nervous and leaning away.
- References: Ayham + Trevor refs + Trevor’s trailer/Sandy Shores screenshot
- Best cut: ~2 sec: Trevor leans in/gestures → Ayham recoils
- Audio: No dialogue. Desert wind, breath, background crackle/smoke.
A. Final still: Nano Banana Pro
Create a cinematic photorealistic live-action Grand Theft Auto V scene using the provided reference images of Ayham and Trevor Philips. Preserve Ayham’s exact facial identity, bald head, beard, skin tone, age, and body proportions from his reference. Preserve Trevor Philips’ exact recognizable facial identity, heavily receding hairline, messy thinning hair, stubble, lean build, and facial structure from his reference. Use any provided GTA V screenshot of Trevor’s trailer only as an environment reference.
Scene: Ayham and Trevor are standing outside Trevor’s iconic trailer in Sandy Shores immediately after some completely insane off-screen chaos. Both are dusty, dirty, and slightly disheveled, but their emotions are completely different.
Trevor is furious and visibly losing his temper. He is leaning slightly toward Ayham with an intense angry expression, eyes wide, jaw tense, one hand gesturing aggressively as if shouting, “What the hell did you do?” His body language should feel unpredictable and intimidating, like he could explode at any second, while still retaining Trevor’s darkly comedic GTA personality.
Ayham is standing beside him looking genuinely nervous and slightly scared, leaning subtly away from Trevor with raised eyebrows and a tense awkward expression, like he is realizing this situation has gone way too far. His hands should be positioned naturally in a defensive or uncertain way, as if trying to calm Trevor down without making him even angrier. The contrast between Trevor’s rage and Ayham’s nervous reaction should make the scene funny.
Dress Ayham in rugged desert-appropriate clothing: a dusty faded dark charcoal T-shirt, open olive-brown overshirt, worn dark jeans, and rugged boots. His clothing should look dirty from the recent chaos. Dress Trevor in his recognizable dirty off-white T-shirt, faded jeans, and worn boots, also covered in dust and grime.
The setting is directly outside Trevor’s run-down Sandy Shores trailer. Surround the trailer with dry desert terrain, junk, scrap metal, utility poles, old equipment, scattered debris, and dusty roadside clutter. In the background, show evidence of whatever caused Trevor’s anger: a small damaged vehicle or broken object, light smoke rising from something, scattered debris, and a faint small fire in the distance. Keep the destruction secondary to the characters.
Camera composition: cinematic medium-wide vertical shot showing Ayham and Trevor from approximately the knees or thighs upward. Trevor is slightly closer to Ayham, invading his personal space, while Ayham subtly leans away. Both faces must be large, clear, and immediately readable. Trevor’s anger and Ayham’s nervous fear are the primary focus, with the trailer clearly recognizable behind them.
Lighting: warm harsh late-afternoon Sandy Shores sunlight, dusty golden atmosphere, realistic shadows, subtle desert haze, and cinematic depth of field.
Visual style: photorealistic cinematic live-action adaptation of Grand Theft Auto V, gritty desert realism, high-budget crime-comedy movie cinematography, believable real humans, realistic skin, dust, clothing, trailer materials, and subtle GTA-inspired color grading. The scene should feel tense but darkly funny, like a real moment from a live-action GTA V movie. Not gameplay footage, not cartoon, not illustration, not cel-shaded.
Vertical 9:16 composition for TikTok, Instagram Reels, and YouTube Shorts. No subtitles, no text, no HUD, no logos, no watermark, no duplicate characters, no distorted faces, no malformed hands, and no extra limbs.
B. Motion: Seedance 2.5
Animate the provided image into an 8-second photorealistic cinematic live-action shot. Preserve Ayham and Trevor’s exact facial identities, clothing, body proportions, trailer, Sandy Shores environment, lighting, and composition.
Trevor becomes increasingly irritated with Ayham. He leans slightly closer with an intense angry expression and makes sharp frustrated hand gestures while staring directly at Ayham. His movements should feel unpredictable and unmistakably Trevor, but he does not physically attack Ayham.
Ayham becomes visibly nervous. He subtly leans backward away from Trevor, his eyebrows rise, his eyes widen slightly, and he lifts his hands in a small confused defensive gesture as if thinking, “What did I do?”
Dust moves through the desert air, light smoke drifts naturally behind them, and their clothing shifts slightly in the breeze. Add blinking, breathing, small head movements, realistic facial micro-expressions, and a slow cinematic camera push toward them.
No spoken dialogue and no lip-synced talking. Trevor may make subtle frustrated breathing or nonverbal sounds only. Use desert wind, distant crackling/fire ambience, and environmental sounds if needed.
No hitting, no weapons, no new characters, no exaggerated screaming, no identity changes, no facial distortion.
SCENE 05: Heist Planning: The Whole Crew
The shift from individual memories to “we were a crew.” Michael leads.
- References: Ayham + Michael + Franklin + Trevor refs + GTA heist-planning screenshot
- Best cut: 2.5-3 sec: Michael points/explains → Franklin nods → group reacts
- Audio: Latest choice: Michael can speak naturally. Others do not speak; they nod/react.
A. Final still: Nano Banana Pro
Create a cinematic photorealistic live-action Grand Theft Auto V scene using the provided reference images of Ayham, Michael De Santa, Trevor Philips, and Franklin Clinton. Preserve each character’s exact recognizable facial identity, age, skin tone, hairstyle or hairline, beard/stubble, body proportions, and overall appearance from their reference images. Use the provided GTA V heist-planning screenshot only as an environment, layout, and atmosphere reference, recreating it as a believable photorealistic live-action version rather than game graphics.
Scene: Ayham, Michael, Trevor, and Franklin are all together inside a classic GTA-style heist planning room, preparing for one last major robbery. A large planning board fills the wall behind them, covered with maps, routes, diagrams, surveillance images, equipment notes, and robbery plans. The room should instantly feel like a real-life version of a GTA V heist setup.
Michael stands closest to the planning board and is clearly leading the meeting. He is pointing at part of the board with a serious, focused, slightly stressed expression, like he is explaining the plan and making sure everyone understands it.
Franklin stands nearby, paying close attention to the board with a calm, focused, professional expression. He should look engaged and serious, like the reliable member of the crew.
Trevor stands on the other side with restless, chaotic energy. He should look impatient, slightly irritated, and unpredictable, like he is already tired of standing around and just wants to get on with the action.
Ayham stands with the group as a full member of the crew, looking at the board with a subtle confident smile, like he knows this plan is crazy but he is fully in. His expression should feel relaxed and slightly amused compared to Michael’s seriousness and Trevor’s instability.
Dress all four characters in stylish heist-planning outfits that fit their personalities. Ayham wears a fitted black dress shirt with sleeves slightly rolled up, dark tailored trousers, dark shoes, and a premium watch. Michael wears a dark charcoal or navy suit with an open-collar dress shirt and dark dress shoes, looking polished and expensive. Franklin wears smart premium streetwear, such as a dark fitted jacket or bomber over a clean shirt, dark trousers or jeans, and fresh sneakers, looking sharp but practical. Trevor wears rugged casual clothing that fits his GTA V character, such as a worn jacket or dirty off-white T-shirt with faded jeans and old boots, making him look slightly out of place compared to the others.
The room should include believable heist details: a table nearby with blueprints, photos, gloves, duffel bags, radios, and small tactical gear. Keep these details secondary and do not clutter the image. The planning board should be large and visually important, but the four characters must remain the main focus.
Camera composition: cinematic vertical 9:16 medium-wide shot, showing all four characters clearly in the frame from approximately thigh or waist level upward, while keeping their faces recognizable and the planning board visible behind them. Michael should be closest to the board, with Franklin, Trevor, and Ayham positioned around him in a natural group composition. The scene should feel like a real crew moment, not a posed lineup.
Lighting: moody indoor lighting with soft overhead illumination, subtle practical lights, realistic shadows, cinematic depth of field, and a premium high-budget heist-movie atmosphere.
Visual style: photorealistic cinematic live-action adaptation of Grand Theft Auto V, believable real human characters, realistic skin, clothing, room materials, maps, and equipment, with subtle GTA-inspired color grading. It should look like a frame from a premium live-action GTA V movie, not gameplay footage, not illustration, not cartoon, and not cel-shaded.
Keep all four faces clear and recognizable. No subtitles, no HUD, no logos, no watermark, no duplicate characters, no distorted faces, no malformed hands, no extra limbs, and no messy unreadable composition.
B. Motion: Seedance 2.5
Animate the provided image into an 8-second photorealistic cinematic live-action scene. Preserve Ayham, Michael, Franklin, and Trevor exactly as shown in the source image, including their facial identities, clothing, body proportions, planning room, board, table, gear, lighting, and environment.
The scene must feel natural, active, and alive, like a real crew planning a heist.
Michael is the leader of the meeting and the only character who speaks. He starts by pointing clearly at one section of the planning board, then shifts slightly and points to another part of the plan while naturally explaining it. His arm, shoulders, head, and eyes should all move in a believable way as he talks through the strategy. His lip sync should look natural and match his speaking.
Franklin listens carefully, following Michael’s hand and gaze from one part of the board to the next. He gives a natural nod of understanding and slightly shifts his posture, like he is focused and following the plan.
Trevor looks impatient and restless. He shifts his weight, briefly looks away in frustration, then looks back toward Michael with an irritated expression. He does not speak, but gives natural reactions like a small exhale, jaw tension, or a reluctant nod.
Ayham stands with folded arms, watching the interaction. He glances between Michael and Trevor, then gives a small restrained amused smile, like he finds the situation a bit funny but is still paying attention. He should also nod slightly at one point.
All characters should have continuous natural motion: blinking, breathing, eye tracking, posture shifts, head movement, small hand movement, realistic facial reactions, and subtle body motion throughout the full shot. Do not let them freeze.
The camera should slowly push in with a slight side drift, creating the feel of a real movie shot. Add very subtle handheld realism so it does not feel static.
Dialogue: Only Michael speaks. His tone should sound calm, serious, and professional, like he is explaining the plan to the crew. The others do not speak and should not have lip-synced dialogue. They only react naturally with nods, facial expressions, and body language.
If audio is generated, include only Michael’s voice plus subtle room ambience, small clothing movement, light footsteps or shifting sounds, and soft indoor environmental sound.
Keep the planning board graphics stable. Do not generate new readable text. No new characters, no camera spins, no morphing faces, no exaggerated gestures, no broken anatomy, and no loss of facial identity.
SCENE 06: Vault: Grabbing the Cash
The payoff. Matching black heist uniforms, masks pulled down, faces visible.
- References: All four character refs + optional vault environment reference
- Best cut: 2.5-3 sec: Franklin loads bag / Trevor grabs stacks / Ayham laughs
- Audio: No dialogue. Cash handling, bags, footsteps, laughter, vault room tone.
A. Final still: Nano Banana Pro
Create a cinematic photorealistic live-action Grand Theft Auto V heist scene using the provided reference images of Ayham, Michael De Santa, Franklin Clinton, and Trevor Philips. Preserve each character’s exact recognizable facial identity, age, skin tone, hairstyle or hairline, beard or stubble, facial proportions, and body structure from their reference images. If a GTA V vault screenshot is provided, use it only as the environment, architecture, and atmosphere reference.
Scene: Ayham, Michael, Franklin, and Trevor are inside a massive high-security bank vault immediately after successfully breaking in. The vault is filled with enormous stacks of bundled cash stored on metal shelves, carts, and tables. Several large black duffel bags are open and being rapidly filled with money.
The whole crew is actively grabbing cash rather than posing for the camera.
Michael is focused and serious, quickly taking organized bundles of money from a shelf and placing them into an open duffel bag, acting like the professional who wants everyone to stay on schedule.
Franklin is grabbing large stacks of cash with both hands and loading them into another bag, smiling confidently and clearly enjoying the success of the robbery.
Trevor is completely overwhelmed with excitement, holding huge bundles of cash in both hands with a wild, almost unbelievable grin, looking like he wants to take every dollar in the vault.
Ayham is standing beside the crew grabbing several thick bundles of cash and stuffing them into an open duffel bag, laughing with adrenaline and disbelief at the enormous amount of money around them.
Dress all four characters in coordinated black heist uniforms, as if they entered the bank together as one professional robbery crew. Ayham, Michael, Franklin, and Trevor should all wear fitted black tactical jackets or long-sleeve tops, black tactical or cargo trousers, black gloves, and rugged black boots. Their uniforms should be similar enough to clearly show they are part of the same crew, while allowing small differences in fit and details to suit each character’s body and personality.
Each character originally wore a black balaclava or ski mask during the robbery, but they have now pulled the masks completely down around their necks after reaching the vault. The balaclavas should remain clearly visible hanging loosely around their necks, but all four faces must be fully uncovered and clearly recognizable. Do not cover their eyes, nose, mouth, forehead, beard, hairline, or any facial features.
Their black uniforms should look slightly worn and disheveled from the break-in, with natural wrinkles, subtle dust, sweat, and signs of physical movement. Keep the clothing realistic, practical, dark, and cinematic rather than futuristic, military-heavy, or overly armored.
The vault should feel enormous, expensive, and highly secure: thick reinforced steel walls, a massive circular open vault door visible behind the group, heavy locking mechanisms, rows of safety-deposit boxes, industrial metal shelving, money carts, overhead security lighting, and countless realistic bundles of cash. The open vault door should help make the location immediately recognizable.
Some cash bundles can be scattered naturally across the floor and tables, but keep the scene believable rather than creating an unrealistic mountain of loose money.
Camera composition: cinematic vertical 9:16 medium-wide shot from slightly inside the vault, positioned low enough that stacks of money and an open duffel bag create foreground depth. All four characters must be clearly visible and actively interacting with the cash. Keep their faces large enough to remain recognizable. The image should feel like a candid moment captured in the middle of the robbery, not a posed group photograph.
Lighting: dramatic cool-white vault lighting mixed with subtle warm highlights on the characters, realistic reflections from the steel surfaces, cinematic shadows, slight atmospheric haze, and high-end movie depth of field.
Visual style: photorealistic cinematic live-action adaptation of Grand Theft Auto V, high-budget heist-movie cinematography, believable real human characters, realistic black tactical clothing, cash bundles, steel surfaces, duffel bags, gloves, masks, and security-vault materials, with subtle GTA-inspired color grading. It should look like a frame from a premium live-action GTA V movie, not gameplay footage, not illustration, not cartoon, and not cel-shaded.
Vertical 9:16 composition for TikTok, Instagram Reels, and YouTube Shorts. All four faces fully visible. Black balaclavas pulled down around their necks only. No guns, no firearms, no weapons. No subtitles, no HUD, no logos, no watermark, no duplicate characters, no distorted faces, no malformed hands, no extra limbs, and no nonsensical duplicated cash.
B. Motion: Seedance 2.5
Animate the provided image into an 8-second photorealistic cinematic fictional heist scene. Preserve Ayham, Michael, Franklin, Trevor, their exact facial identities, black uniforms, gloves, masks resting around their necks, vault environment, duffel bags, money, and composition.
Franklin quickly picks up several existing bundles of cash and places them into the open duffel bag.
Michael remains serious and methodical, taking bundles from the table and organizing them into another bag.
Trevor excitedly gathers several bundles against his chest with a wild delighted grin.
Ayham grabs existing cash bundles and quickly places them into his bag while laughing with adrenaline.
Add visible arm motion, body weight shifts, natural interaction with the existing money, blinking, breathing, clothing movement, and mild handheld camera energy. The camera slowly moves deeper into the vault.
No spoken dialogue. No lip-synced talking. Only natural laughter, cash handling sounds, bags moving, footsteps, vault-room ambience, and subtle environmental sound effects.
Keep all four faces completely uncovered. Masks remain around their necks. No weapons, no alarms, no police, no new people, no flying money, no masks covering faces, no identity changes.
SCENE 07: Money Table: After the Score
Release and celebration. The heist is over; the crew can finally breathe.
- References: All four character refs + GTA money-table/composition screenshot
- Best cut: 2.5-3 sec: Trevor laughs → Ayham joins → Michael small toast
- Audio: No dialogue. Laughter, chair/glass movement, room ambience.
A. Final still: Nano Banana Pro
Create a cinematic photorealistic live-action Grand Theft Auto V scene using the provided reference images of Ayham, Michael De Santa, Franklin Clinton, and Trevor Philips. Preserve each character’s exact recognizable facial identity, age, hairstyle, skin tone, beard or stubble, facial proportions, and overall appearance from their character references. Use the provided GTA screenshot as the primary reference for the composition, conference room, large table, sunset windows, money, and post-heist atmosphere, recreating everything as a believable photorealistic live-action scene rather than game graphics.
Scene: Ayham, Michael, Franklin, and Trevor are sitting together around a large modern conference table immediately after successfully completing a huge heist. The table is covered with large stacks and bundles of cash, several open duffel bags filled with money, loose banknotes, maps, blueprints, radios, gloves, and a few masks. It should instantly look like the crew has just pulled off one of the biggest robberies of their lives.
Michael sits back in his chair looking exhausted but deeply satisfied, holding a drink and giving the group a subtle relieved smile, like he cannot believe the plan actually worked.
Franklin is relaxed and smiling confidently while looking across the table at the others, clearly enjoying the victory.
Trevor looks absolutely thrilled and slightly unhinged, leaning forward toward the piles of cash with a huge excited grin, almost laughing at how much money is sitting in front of them.
Ayham sits naturally with the crew, laughing and looking between Trevor and Michael, clearly enjoying the ridiculous moment. He should feel completely integrated into the group rather than positioned separately as an outsider.
Dress Ayham in a fitted black shirt with sleeves slightly rolled, dark trousers, and a premium watch. Michael wears a dark charcoal suit with his shirt collar slightly open after the heist. Franklin wears sophisticated dark Los Santos streetwear with a premium jacket and clean shirt. Trevor wears his recognizable rough, worn casual clothing, creating a funny visual contrast with the expensive room and enormous amount of money.
The conference room should feel luxurious and slightly criminal: enormous floor-to-ceiling windows overlooking Los Santos during a beautiful orange-pink sunset, dark modern walls, expensive chairs, and a huge dark table. Place the money prominently in the foreground so bundles of cash create depth leading toward the four characters.
Camera composition: cinematic vertical 9:16 shot from one end of the table, slightly low and close to the stacks of money. Large bundles of cash occupy part of the foreground while all four characters are clearly visible around the table beyond them. Their faces must remain large enough to recognize. The composition should feel candid, like the camera captured a private moment between the GTA crew immediately after a successful score.
Lighting: dramatic sunset light entering through the floor-to-ceiling windows, warm orange and pink highlights mixed with subtle indoor lighting, realistic reflections across the table, cinematic shadows, and premium depth of field.
Visual style: photorealistic cinematic live-action adaptation of Grand Theft Auto V, high-budget crime-film cinematography, realistic human characters, realistic cash, clothing, furniture, glass, and city lighting, subtle GTA-inspired color grading. It should look like a real frame from a premium live-action GTA V movie, not gameplay footage, not illustration, not cartoon, and not cel-shaded.
Vertical 9:16 composition for TikTok, Instagram Reels, and YouTube Shorts. No guns, no firearms, no weapons, no subtitles, no HUD, no game interface, no watermark, no duplicate characters, no distorted faces, no malformed hands, no extra limbs, and no nonsensical duplicated money bundles.
B. Motion: Seedance 2.5
Animate the provided image into an 8-second photorealistic cinematic live-action shot. Preserve all four characters’ exact identities, clothing, money, table, maps, room, sunset, and composition.
Trevor leans slightly forward over the table and starts laughing excitedly. Ayham turns toward Trevor and laughs with him. Franklin smiles and looks between them. Michael relaxes deeper into his chair, gives the group a satisfied smile, and slowly lifts his drink slightly in a quiet celebratory gesture.
Add natural shoulder movement from laughter, small head turns, blinking, breathing, hand movement, subtle interaction with the existing cash, and gentle sunset reflections moving across the room.
The camera performs a slow cinematic sideways drift or very subtle push-in.
No spoken dialogue. No lip-synced talking. Only natural laughter, subtle room ambience, glass movement, chair movement, and environmental sounds if generated.
Do not create flying money, new objects, weapons, extra people, facial morphing, malformed hands, or exaggerated reactions.
SCENE 08: Michael: Poolside Peace
The pacing slows. A rare calm moment before the goodbye.
- References: Ayham + Michael refs; mansion/pool environment is already established in the still
- Best cut: 2.5-3 sec: Michael turns/smiles → Ayham laughs softly
- Audio: No dialogue. Pool ambience, breeze, birds, subtle glass movement.
A. Final still: Nano Banana Pro
Create a cinematic photorealistic live-action Grand Theft Auto V scene using the provided reference images of Ayham and Michael De Santa. Preserve both characters’ exact facial identities, age, skin tone, hairstyle, beard, and recognizable facial structure.
Scene: Ayham and Michael are relaxing together beside the swimming pool at Michael’s luxurious Rockford Hills mansion during golden hour. They are sitting comfortably in expensive outdoor lounge chairs close to the edge of the pool, each holding a drink, sharing one of those rare peaceful moments where nothing is exploding and nobody is chasing them.
Michael is leaning back with a relaxed but slightly tired expression, looking toward Ayham with a subtle amused smile. Ayham is looking toward Michael and smiling, as if they have just finished joking about all the ridiculous things they have survived together. Their body language should feel natural, comfortable, and like longtime friends.
Dress Ayham in an upscale relaxed Los Santos outfit: fitted cream or white short-sleeve polo shirt, tailored beige trousers, clean white sneakers, and a simple luxury watch. Dress Michael in an expensive casual outfit that fits his personality: dark navy polo shirt, beige shorts or light trousers, loafers, and a luxury watch.
Michael’s iconic Los Santos mansion should be clearly recognizable in the background, with Mediterranean-style architecture, large windows, manicured landscaping, palm trees, outdoor furniture, and the swimming pool reflecting the warm sunset sky.
Place two realistic glasses with amber-colored drinks and ice in their hands. The atmosphere should feel wealthy, calm, nostalgic, and slightly funny, like an unusually peaceful GTA V afternoon.
Camera composition: cinematic medium-wide vertical shot from slightly across the pool, with Ayham and Michael dominating the foreground. Their faces should be clearly visible, while the pool and mansion create depth behind them. Natural relaxed poses, no staged-looking posture.
Lighting: beautiful warm California golden-hour sunlight, soft orange highlights on their faces, reflections across the swimming pool, long gentle shadows, realistic skin tones, cinematic depth of field.
Visual style: photorealistic cinematic live-action adaptation of Grand Theft Auto V, premium high-budget crime-comedy movie look, realistic humans, subtle GTA-inspired color grading, natural skin and clothing textures, realistic architecture and water. It should feel like a real moment inside a live-action GTA V movie, not gameplay, not illustration, not cartoon, and not cel-shaded.
Vertical 9:16 composition. No subtitles, no text, no HUD, no logos, no watermark, no distorted faces, no malformed hands, no duplicate characters or objects.
B. Motion: Seedance 2.5
Animate the provided image into an 8-second photorealistic cinematic live-action shot. Preserve Ayham, Michael, their exact identities, outfits, lounge chairs, drinks, pool, mansion, sunset, and composition.
Michael turns his head slightly toward Ayham and gives him a knowing amused look. Ayham looks toward Michael and starts smiling, then gives a small natural laugh. Michael responds with a subtle grin.
Both remain comfortably seated and relaxed. They casually adjust their drinks slightly in their hands without drinking from them. Their body language should feel like two longtime friends enjoying a rare peaceful moment.
The swimming pool gently ripples and reflects the sunset. Palm leaves move slightly in the breeze. Warm golden light shifts subtly across their faces, clothing, and the water.
Add natural blinking, breathing, tiny facial reactions, hand movement, water movement, and a very slow cinematic lateral camera drift.
No spoken dialogue and no lip-synced talking. Only a subtle natural laugh, pool ambience, light breeze, distant birds, glass movement, and environmental sound effects if audio is generated.
No sudden movement, no new characters, no spilling drinks, no facial morphing, no camera spin.
SCENE 09: Vinewood: Final GTA V Goodbye
The emotional centerpiece. Ayham is not happy; the others quietly accept the goodbye.
- References: Ayham + Michael + Franklin + Trevor refs + Vinewood screenshot
- Best cut: 4-5 sec: Ayham looks at each of them → they turn toward Los Santos
- Audio: No dialogue. Wind, distant city, birds. Let the music carry emotion.
A. Final still: Nano Banana Pro
Create a cinematic photorealistic live-action Grand Theft Auto V final goodbye scene using the provided reference images of Ayham, Michael De Santa, Franklin Clinton, and Trevor Philips. Preserve each character’s exact recognizable facial identity, age, skin tone, hairstyle or hairline, beard or stubble, facial proportions, and body appearance from their references.
Use the provided Vinewood sign screenshot as the primary environment and location reference. Recreate the same iconic Vinewood hills setting as a believable photorealistic live-action location, keeping the large VINEWOOD sign, dry California hills, communication towers, scattered trees, dirt paths, and recognizable Los Santos landscape. Transform the environment from game graphics into realistic cinematic photography while preserving the identity of the location.
Scene: Ayham, Michael, Franklin, and Trevor are standing together on the hillside near the Vinewood sign during a beautiful sunset. This is their final moment together after all the chaos, heists, fights, parties, and insane memories they have shared. Nothing dangerous is happening. There are no police, no action, and no chaos. Everything is peaceful and unusually quiet.
Position all four characters naturally beside each other overlooking Los Santos, with the Vinewood sign clearly visible behind or beside them. They should feel like longtime friends sharing one final memory together rather than posing for a photograph.
Ayham stands among Michael, Franklin, and Trevor, visibly emotional and unusually quiet. He is not smiling. His expression is serious, heavy, and bittersweet, with slightly tense eyes and a restrained sadness, like he is trying to hold himself together while realizing this may be the last time they are all together. He occasionally looks from Michael to Franklin to Trevor, taking in their faces one last time. His emotion should communicate gratitude, attachment, and genuine difficulty saying goodbye without exaggerated crying or melodrama.
Michael looks unusually peaceful and reflective. He gives Ayham a very small bittersweet smile, as though he understands exactly what Ayham is feeling but does not need to say anything.
Franklin has a calm, warm expression and looks toward Ayham with quiet understanding. His subtle smile should feel reassuring rather than happy.
Trevor, for once, is completely calm. He looks toward Ayham with an unusually sincere, almost vulnerable expression and only the faintest bittersweet half-smile. His usual chaos is completely absent.
Their interaction should feel like a genuine farewell between four people who have been through everything together. Ayham should clearly be the most emotional person in the scene. He should not laugh or smile. Use lingering eye contact, restrained facial emotion, relaxed but heavy body language, and perhaps Michael or Franklin placing a hand naturally on Ayham’s shoulder. Nobody looks toward the camera. The moment should feel private, uncomfortable in an emotional way, and difficult to end.
Dress Ayham in a clean understated cinematic outfit: fitted dark charcoal T-shirt, lightweight black jacket, dark jeans or trousers, clean sneakers, and a simple watch. Dress Michael in sophisticated relaxed clothing: dark navy jacket or blazer over an open-collar light shirt, dark trousers, and understated luxury shoes. Dress Franklin in mature premium Los Santos streetwear: dark green or black jacket over a clean shirt, dark trousers, and fresh sneakers. Dress Trevor in his recognizable rugged style: worn faded T-shirt, old dark jacket, faded jeans, and rugged boots. Keep him rough and authentic rather than polished.
The sun is setting behind the Los Santos hills, filling the sky with rich orange, gold, pink, and subtle purple tones. Warm sunset light should wrap around the four characters, creating beautiful rim lighting around their shoulders and faces. The Vinewood sign catches the final warm sunlight while Los Santos is visible far below in atmospheric haze.
Camera composition: cinematic vertical 9:16 medium-wide shot, positioned slightly below or level with the characters so the group feels heroic but intimate. Show all four characters clearly from roughly the knees or thighs upward while still revealing the iconic Vinewood sign and sunset environment behind them. Keep every face large enough to remain clearly recognizable.
The composition should initially show the characters interacting with each other, but also allow them to naturally turn and look out toward Los Santos together, creating the feeling of saying goodbye to the city.
Lighting: beautiful golden-hour sunset cinematography, realistic warm rim light, soft facial illumination, long natural shadows, atmospheric haze, subtle lens flare, realistic skin tones, and premium cinematic depth of field.
Visual style: photorealistic cinematic live-action adaptation of Grand Theft Auto V, high-budget emotional movie cinematography, believable real human characters, realistic hills, vegetation, communication towers, Vinewood sign materials, clothing, and natural California sunset lighting. Subtle GTA-inspired cinematic color grading, nostalgic, bittersweet, and emotional without becoming overly dramatic. It should feel like the final scene of a premium live-action GTA V movie.
Vertical 9:16 composition for TikTok, Instagram Reels, and YouTube Shorts. All four characters must be present and recognizable. The Vinewood sign must be clearly visible. No cars, no police, no chase, no weapons, no subtitles, no HUD, no logos, no watermark, no extra people, no duplicate characters, no distorted faces, no malformed hands, and no extra limbs.
B. Motion: Seedance 2.5
Animate the provided image into an 8-second emotional photorealistic cinematic live-action shot. Preserve Ayham, Michael, Franklin, Trevor, their exact identities, clothing, positions, Vinewood sign, hillside, sunset, and composition.
Ayham is the emotional center of the scene. He does not smile. He slowly looks toward Michael, then Franklin, then Trevor, taking in each of them one final time. His expression remains serious, restrained, and visibly emotional without crying dramatically.
Michael keeps his hand gently on Ayham’s shoulder and gives him a small reassuring look.
Franklin watches Ayham with a calm, warm expression and gives the slightest nod.
Trevor remains unusually quiet and sincere, holding Ayham’s gaze briefly with a restrained bittersweet expression.
Near the end of the shot, the group slowly begins turning their attention toward Los Santos and the Vinewood hills, creating the feeling that the goodbye is becoming real.
Add subtle breathing, blinking, tiny facial reactions, a light breeze moving their clothing, and gentle golden-hour atmospheric movement. Use an extremely slow cinematic push-in.
Absolutely no spoken dialogue. No lip-synced talking. No laughter. Audio should be limited to soft wind, distant city ambience, birds, and subtle emotional environmental sound if generated.
Keep the performance restrained and realistic. No dramatic gestures, no hugging animation unless naturally initiated from the existing pose, no identity changes.
SCENE 10: GTA VI: New Crew, New Mission
The handoff. Jason + Lucia welcome Ayham; Lucia gives him the RS7 keys.
- References: Ayham + Jason + Lucia uploaded refs. Character references override all styling assumptions.
- Best cut: 8-10 sec: Lucia asks → pause → Ayham answers → key handoff
- Audio: Dialogue only here: Lucia “Are you ready?” / Ayham “After 13 years… yeah. I’m ready.” Jason stays silent.
A. Final still: Nano Banana Pro
Create a cinematic photorealistic live-action Grand Theft Auto VI transition scene using the provided reference images of Ayham, Jason, and Lucia.
IMPORTANT: The uploaded character reference images are the absolute source of truth for each character’s appearance. Copy each character exactly as shown in their reference image. Do not redesign or reinterpret their clothing, hairstyle, accessories, body type, facial hair, tattoos, jewelry, sunglasses, or overall styling. If a character is shirtless in the reference, keep them shirtless. If they are wearing sunglasses, jewelry, specific pants, shoes, or accessories, preserve those exact details.
Preserve Ayham’s exact facial identity, bald head, beard, skin tone, age, facial proportions, body structure, clothing, and accessories exactly as shown in his uploaded reference.
Preserve Jason’s exact facial identity, hairstyle, facial structure, skin tone, age, muscular build, clothing or lack of clothing, sunglasses, tattoos, jewelry, pants, footwear, and every other visible styling detail exactly as shown in his uploaded reference.
Preserve Lucia’s exact facial identity, hairstyle, skin tone, age, facial proportions, body shape, clothing, jewelry, footwear, accessories, and every other visible styling detail exactly as shown in her uploaded reference.
Do not create new outfits for any of the three characters. The reference images override any assumptions about GTA VI clothing or styling.
Scene: Jason and Lucia have just arrived in Vice City to pick Ayham up for a new mission. Instead of standing apart, they are warmly welcoming him into the new crew.
Ayham stands in the center of the frame. Jason is on Ayham’s left side and naturally wraps one arm around Ayham’s shoulder in a friendly side hug. Lucia is on Ayham’s right side and also embraces him naturally from the side. The interaction should feel warm, confident, and spontaneous, like Jason and Lucia are welcoming Ayham into their next adventure.
Lucia uses her free hand to hand Ayham the car key fob. The key exchange must be clearly visible and easy to understand. Lucia looks toward Ayham with a confident playful smile.
Jason keeps one arm around Ayham and has a relaxed, confident grin, looking ready for whatever mission comes next.
Ayham is still slightly emotional after saying goodbye to Michael, Franklin, and Trevor, but his expression is beginning to change into surprise, acceptance, and a restrained hopeful smile. He looks toward Lucia and the keys as he realizes a new chapter is beginning.
Behind the three characters, prominently show a sleek black Audi RS7 parked at a cinematic three-quarter angle, ready to leave. The car should be clearly recognizable as a modern high-performance luxury sportback with glossy black paint, aggressive headlights, wide stance, large wheels, realistic glass, natural reflections, and premium detailing. One front door may be open.
The environment should strongly evoke modern Vice City at sunset: tall palm trees, tropical roadside architecture, neon storefront signs beginning to illuminate, power lines, humid atmospheric haze, distant city lights, and a beautiful Florida sunset with rich pink, orange, purple, and blue tones.
Camera composition: cinematic vertical 9:16 medium-wide shot. Ayham is centered, Jason hugs him from the left, and Lucia hugs him from the right while clearly handing him the key fob. Their faces and upper bodies should dominate the frame while leaving enough space behind them to clearly show the black Audi RS7.
The image should immediately communicate: Jason and Lucia have arrived, they welcome Ayham into their crew, Lucia gives Ayham the keys, and the three of them are about to leave together for a new mission.
Lighting: premium cinematic tropical sunset lighting, warm rim light around all three characters, soft pink and purple ambient light from the sky, subtle neon reflections, realistic highlights across the RS7, natural skin tones, soft atmospheric haze, and cinematic depth of field.
Visual style: extremely photorealistic cinematic live-action adaptation of Grand Theft Auto VI, realistic human skin, natural expressions, realistic hair, fabric, jewelry, sunglasses, tattoos, and vehicle materials, with subtle GTA-inspired cinematic color grading. It should feel like a frame from a high-budget live-action GTA VI movie.
Most important rule: match Ayham, Jason, and Lucia to their uploaded reference images exactly. Do not change their clothing, accessories, hairstyles, body appearance, or styling. Character-reference accuracy takes priority over all written styling assumptions.
Exactly three main characters: Ayham, Jason, and Lucia. No additional people, no weapons, no police, no subtitles, no HUD, no watermark, no duplicate characters, no extra arms, no malformed hands, no distorted faces, and no duplicated keys.
B. Motion: Seedance 2.5
Animate the provided image into an 8–10 second photorealistic cinematic live-action scene. Preserve Ayham, Jason, and Lucia exactly as shown in the source image, including their facial identities, clothing, accessories, sunglasses, hairstyles, body appearance, positions, and the black Audi RS7 behind them. Do not redesign any character.
Jason keeps one arm naturally around Ayham’s shoulder in a friendly welcoming side hug. Lucia remains close on Ayham’s other side and slowly extends the car key fob toward Ayham.
Ayham initially looks down at the key with a serious, emotional expression, still processing the goodbye he has just experienced.
Lucia looks directly toward Ayham and naturally asks: “Are you ready?”
Ayham pauses for a brief moment. He looks at Lucia, then toward Jason, and a small emotional smile slowly appears. Ayham replies naturally: “After 13 years… yeah. I’m ready.”
Lucia places the key fob naturally into Ayham’s hand. Jason gives Ayham a small friendly squeeze on the shoulder. After Ayham’s response, all three share a subtle confident smile, like the next adventure is about to begin.
Dialogue must sound natural, restrained, and cinematic, not theatrical. Lip synchronization must match only the character currently speaking. Jason does not speak.
Add subtle breathing, blinking, natural hand movement during the key exchange, slight body movement from the side hugs, gentle tropical breeze through hair and clothing, and realistic sunset reflections moving across the Audi RS7.
Camera performs a very slow cinematic push-in. Keep the moment warm, emotional, and grounded.
Only Lucia and Ayham speak. No other dialogue, background voices, narration, or additional speech.
No weapons, no police, no new people, no exaggerated hugging, no facial morphing, no duplicate key, no extra fingers or arms.
03 Final edit timeline
The order that gives the montage an emotional arc instead of feeling like random GTA clips.
# · Scene · Cut · Beat · Why
01 · Trevor, Truck Chaos · 2-2.5s · HOOK · Start with the strongest movement and instant GTA chaos.
02 · Franklin, Club Exit · 2-2.5s · FUN · Personal friendship beat; keep the energy up.
03 · Michael, Tennis · 2-2.5s · COMEDY · Michael side-eye + Ayham laugh.
04 · Trevor, Trailer Argument · ~2s · CHAOS · Very quick dark-comedy beat.
05 · Heist Planning · 2.5-3s · CREW · Shift from individual memories to the full crew.
06 · Vault, Cash · 2.5-3s · PAYOFF · Fast movement, bags and cash.
07 · Money Table · 2.5-3s · RELEASE · The crew celebrates and breathes.
08 · Michael, Poolside · 2.5-3s · SLOW DOWN · Signal the emotional turn.
09 · Vinewood Goodbye · 4-5s · GOODBYE · Let this breathe. Ayham is serious, not smiling.
10 · GTA VI, Jason + Lucia · 8-10s · NEW BEGINNING · Dialogue + keys. This is the payoff/end.
ARC: CHAOS → FUN → COMEDY → CHAOS → CREW → HEIST → SUCCESS → PEACE → GOODBYE → NEW BEGINNING.
TARGET LENGTH: Roughly 30-35 seconds. The first seven scenes should feel like memories flashing by. Poolside begins the slowdown. Vinewood and GTA VI get the most breathing room.
04 Audio & dialogue rules
Keep the memories visual. Dialogue is the exception, not the default.
Scene · Voice · Audio
Truck · No speech · Engine + dirt + laughter + distant siren
Club · No speech · Laughter + footsteps + nightlife ambience
Tennis · No speech · Small laugh + court ambience
Trevor trailer · No speech · Wind + breath + faint crackle
Planning · Michael only · Michael explains the plan; everyone else nods/reacts
Vault · No speech · Cash + bags + footsteps + laughter
Money table · No speech · Laughter + glass/chair movement
Pool · No speech · Pool + breeze + birds
Vinewood · No speech · Wind + distant city + emotional music
GTA VI ending · Lucia + Ayham · Lucia: “Are you ready?” Ayham: “After 13 years… yeah. I’m ready.”
13-YEAR LINE: If the post goes live before September 17, 2026, “Almost 13 years…” is technically precise. On/after September 17, “After 13 years…” is exact.
06 Production checklist
A fast repeatable checklist if you remake the concept or hand it to an editor.
1. Choose 9:16 from the beginning.
2. Upload identity references first; use GTA screenshots/art only for environment/prop/composition.
3. Generate the final cinematic still before touching Seedance.
4. Inspect face likeness, hands, props, clothing, location, lighting and framing.
5. For each Seedance clip, preserve the still and describe motion only.
6. Generate 8 seconds and cut the best 2-3 seconds. Use 8-10 seconds for the GTA VI dialogue ending.
7. Avoid static prompts: give every person a specific action/reaction and give the camera a subtle move.
8. Keep most clips dialogue-free. Let SFX and music carry the montage.
9. Slow the edit at Poolside, then let Vinewood breathe.
10. End with Jason + Lucia, key handoff, and the final line.
More guides
All guides · All AI tools · Prompts · HandlyTech
