SpaceSkills AI Filmmaking

Tokyo Night Run for Last Train

Vlog / Daily Life Live-action Cinematic 2026-08-30

Tokyo Night Run for Last Train preview

Footage Β© @codewithhajra Β· View original post

The clip was published publicly by its creator and is linked here directly; rights holders may request removal.

Full prompt

Tokyo at night was worth the rush. πŸŒƒ

Created in Minimax H3

Prompt:
FORMAT: 15 seconds | 9:16 vertical | Ultra-photorealistic smartphone vlog | Real-world Tokyo at night

CHARACTER LOCK
Use the attached girl as the exact character and identity reference throughout the entire video. Preserve her recognizable facial features, face shape, warm skin tone, brown eyes, natural makeup, soft features, and dark brown hair styled in a messy high bun with loose strands framing her face.
She is an adult woman with a natural, approachable travel-vlogger appearance.

OUTFIT LOCK

Match the outfit shown in the storyboard:

Cream-colored knit top

Light beige lightweight jacket, worn open

Charcoal/dark gray straight-leg trousers

Clean white sneakers
Small silver hoop earrings
Simple silver wristwatch
Minimal accessories
Natural, understated makeup
The hairstyle, face, outfit, accessories and overall appearance must remain identical across all six storyboard scenes.
00–04s β€” RUNNING FOR IT
Handheld selfie footage on a crowded Tokyo street at night.

The girl from the attached reference walks quickly toward the camera, slightly out of breath. Her beige jacket moves naturally as she walks.

Bright Tokyo signs, storefronts, pedestrians and traffic create an authentic nighttime city background.

She looks directly into the phone camera and laughs:

β€œWaitβ€”my train leaves in two minutes!”

She quickly turns the phone forward and starts jogging toward the station.

04–08s β€” THE STATION

Handheld camera follows her from behind and slightly beside her as she rushes down the station stairs.

Her messy brown bun and beige jacket remain clearly recognizable.

She reaches the ticket gate, taps her transit card and quickly passes through.

The phone camera shakes naturally from her movement and briefly loses autofocus.

08–12s β€” MADE IT

She reaches the platform just as the train arrives.

The camera captures the approaching train headlights before returning to her face.

She steps inside, turns the phone toward herself and smiles while catching her breath.

She says:

β€œI actually made it!”

The train doors close behind her.

12–15s β€” TOKYO THROUGH THE WINDOW

Inside the train, she turns the phone toward the window.

Tokyo's illuminated buildings and streets naturally blur past as the train accelerates.

Her face appears softly reflected in the glass.

She looks toward the city and quietly says:

β€œOkay… Tokyo at night hits different.”

Natural cut.

REALISM & CHARACTER CONSISTENCY

Same attached girl in every shot. No character changes.

Maintain the exact:

Facial identity

Brown eyes

Dark brown messy high bun

Loose face-framing hair strands
Skin tone
Facial proportions
Cream knit top
Beige jacket
Charcoal trousers
White sneakers
Silver earrings
Wristwatch
Natural skin texture, realistic hair strands, authentic blinking, subtle facial expressions and believable body movement.
No outfit changes, no hairstyle changes, no different woman, no face morphing, no beauty-filter effect.

VISUAL STYLE
Authentic handheld smartphone footage rather than a polished movie.

Natural motion blur, autofocus hunting, slight exposure shifts, realistic low-light phone noise, imperfect framing, natural Tokyo crowds and physically accurate movement.

No cinematic transitions, no fantasy effects, no exaggerated neon, no artificial-looking skin, no duplicated people, no impossible camera movement.

AUDIO
Natural Tokyo ambience, footsteps, station announcements, crowd noise, train brakes, door chime and her real voice.

No background music, no subtitles, no watermark.

Details

Target model MiniMax H3
Use case Vlog / Daily Life
Style Live-action Cinematic
Media Video
Duration 16.5s
Aspect ratio 45:56
Likes 110
Published 2026-08-30

Attribution

Author: @codewithhajra Β· View original post

Prompts remain the copyright of their authors. We index and quote them with attribution plus a link to the original post; rights holders may request removal.

Run this prompt through the API

Copy the prompt above into a MiniMax H3 (Hailuo) text-to-video call to reproduce the shot. When adapting it, swap subject, setting and camera move one clause at a time. Access and licensing details are on About.

← Back to all prompts