AI Video Character Consistency: 9 Copy-Paste Multi-Shot Prompts to Keep the Same Face

Nothing kills an AI short drama faster than a lead with a different face in every shot. Character consistency is the number-one complaint of serial-fiction creators using Veo 3.1, Kling 4.0, Seedance 2.5 or Hailuo 3 — and the fix is three layers of discipline, not one magic feature. Below: the system plus 9 copy-paste prompts.

Why Your Character’s Face Keeps Changing

• Each clip is sampled independently — every generation re-rolls “young woman with black hair” from scratch, and text underdetermines a face.
• You paraphrase between shots. Change “long black hair” to “dark hair” and you’ve handed the sampler a different person.
• Angles, lighting and wide shots expose drift first — a face that holds in profile can melt the moment you cut in.

The Formula: One Locked Character Block

text [CHARACTER LOCK — copy this block into every shot prompt, unchanged] A 28-year-old East Asian woman, oval face, almond-shaped dark brown eyes, small mole under her left eye, shoulder-length straight black hair with a side-swept fringe, cream trench coat over a white turtleneck, silver watch on her left wrist.

Three rules:

1. Paste it verbatim — same words, same order, every shot.
2. Lock 2–3 anchors that survive every angle — a mole, scar, glasses, exact hair color, one distinctive garment. Unique beats pretty.
3. Identity first, action second. Character block on top, then action, camera, lighting.

Layer 2: Reference Images (Verified Features)

Text alone pins identity only so far:

• Veo 3.1 — Ingredients to Video: up to 3 reference images (character, object, scene/style) per generation, built to hold characters across shots.
• Kling 4.0 — Omni Reference: up to 15 multimodal references per request (images plus up to 5 video clips) for character continuity.
• Seedance 2.5 — mixed reference sets: up to 30 images, 10 video clips, 10 audio files per generation, cited as [Image1], [Vid1]… — its guidance: reuse the same identity image every generation.
• Hailuo 3 (MiniMax H3) — Omni-Reference: up to 9 reference images plus video and audio references to pin identity across shots.

Two rules: reuse the same identity set every time (face + full-body shot), and use original characters — models block real people and celebrity lookalikes.

Layer 3: Shots That Dodge Drift

• Finish the action inside one shot — fewer cuts, fewer re-rolls.
• Cut away from the face for transitions: hands, objects, back view, silhouette, over-the-shoulder.
• Match eyeline, key light and lens distance across the shot list — and generate the hero close-up first, the anchor for every other shot.

9 Copy-Paste Multi-Shot Prompts

Each group reuses its character block word for word — copy it with every shot.

Group A — Maya: The Rainy Alley (3 shots)

A1. Wide Establishing

text Character (identical in every shot): a 28-year-old East Asian woman, oval face, almond-shaped dark brown eyes, small mole under her left eye, shoulder- length straight black hair with a side-swept fringe, cream trench coat over a white turtleneck, silver watch on her left wrist. Wide shot: she walks down a rain-soaked alley at night, stops under a red paper lantern and looks up at the rain, neon reflections on wet asphalt, slow push-in, cinematic 35mm look, gentle rain audio, bittersweet mood.

A2. Close-Up Reaction

text Character (identical in every shot): a 28-year-old East Asian woman, oval face, almond-shaped dark brown eyes, small mole under her left eye, shoulder- length straight black hair with a side-swept fringe, cream trench coat over a white turtleneck, silver watch on her left wrist. Medium close-up: she stands under the red paper lantern, takes out her phone, reads the screen and smiles softly, warm lantern key light, cool neon rim light, shallow depth of field, subtle handheld drift, soft rain audio, quiet hopeful mood.

A3. Over-Shoulder Exit

text Character (identical in every shot): a 28-year-old East Asian woman, oval face, almond-shaped dark brown eyes, small mole under her left eye, shoulder- length straight black hair with a side-swept fringe, cream trench coat over a white turtleneck, silver watch on her left wrist. Over-the-shoulder shot from behind her as she pushes open a noodle-shop door, a bell chimes, warm steam and light spill across her face, she glances back at the alley, camera follows two steps inside, handheld, door chime and kitchen ambience audio.

Group B — Daniel: The Night Detective (2 shots)

B1. Evidence Under Sodium Light

text Character (identical in every shot): a 40-year-old Black man, shaved head, strong jawline, dark brown eyes, short greying beard, thin scar through his right eyebrow, tall athletic build, charcoal wool overcoat over a navy shirt, black leather gloves. Interior medium shot in a parked car at night: he opens a manila evidence envelope under a warm sodium parking-garage light, photographs slide onto his gloved palm, slow push-in from the passenger side, orange lamp glow through the windshield, tense mood, muffled rain audio.

B2. The Approach

text Character (identical in every shot): a 40-year-old Black man, shaved head, strong jawline, dark brown eyes, short greying beard, thin scar through his right eyebrow, tall athletic build, charcoal wool overcoat over a navy shirt, black leather gloves. Wide shot: he steps out of the car, closes the door and walks toward a brick entrance, silhouetted against warm doorway light, static low-angle camera, wet street reflections, cold blue night grade, footsteps audio.

Group C — Sofia: The Kitchen Host (2 shots)

C1. The Espresso Demo

text Character (identical in every shot): a 32-year-old Latina woman, warm olive skin, round face, curly shoulder-length dark brown hair tied half-up, hazel eyes, gold hoop earrings, sage green linen apron over a white tee. Medium shot in a sunlit kitchen: she places a cup under a stainless espresso machine, pulls the lever, espresso streams out and steam curls upward, slow push-in over the counter, bright morning window light, cheerful mood, brewing sounds audio.

C2. The First Sip

text Character (identical in every shot): a 32-year-old Latina woman, warm olive skin, round face, curly shoulder-length dark brown hair tied half-up, hazel eyes, gold hoop earrings, sage green linen apron over a white tee. Close-up: she lifts the cup, inhales the aroma, closes her eyes and smiles, soft morning light across her face, shallow depth of field, static camera, warm inviting mood, kitchen ambience audio.

Group D — Leo: The Morning Runner (2 shots)

D1. Bridge Run

text Character (identical in every shot): a 24-year-old white man, wavy sandy-blond hair, blue eyes, light stubble, lean runner's build, navy running jacket over a grey tee, black shorts, blue wireless earbuds. Wide shot: he jogs across a suspension bridge at sunrise, camera tracking alongside at matching pace, long shadows, warm golden backlight over the river, breath vapor in cool air, energetic calm mood, footsteps and wind audio.

D2. Silhouette Cool-Down

text Character (identical in every shot): a 24-year-old white man, wavy sandy-blond hair, blue eyes, light stubble, lean runner's build, navy running jacket over a grey tee, black shorts, blue wireless earbuds. He stops at a park fountain, drinks and catches his breath, backlit by low sun so his face reads mostly in silhouette, camera arcs behind his shoulder toward the skyline, warm rim light, reflective mood.

The Fixed-Field Character Template

Fill one per character; fields are your checklist.

text CHARACTER LOCK (copy into every shot; keep field order and wording identical): AGE + ETHNICITY: FACE SHAPE / EYES / NOSE / BROWS / MOUTH: DISTINGUISHING MARKS (must survive any angle — mole, scar, freckles, glasses): HAIR (color, length, texture, style): BUILD + HEIGHT IMPRESSION: WARDROBE (exact garments, colors, layers): ACCESSORIES (with side — watch on left wrist, ring on right hand): VOICE (only if dialogue/audio matters):

3 Extra Consistency Tricks

1. Version your reference files (CHAR_MAYA_v1_face.png); never mix versions mid-series.
2. Light faces the same way — one key-light direction keeps the geometry reading as one person.
3. Keep units short — 15–20 second units hold identity far better than long ones.

FAQ

Why does drift happen even with the exact same prompt? Independent sampling — even identical text rolls a different face. Fix it with anchors, the lock block, one reference set.

Which model is best for character consistency? All four ship reference systems: Kling 4.0 Omni Reference (15 refs), Seedance 2.5, Veo 3.1 Ingredients (3 images), Hailuo 3 Omni-Reference (9 images).

Can I use a photo of a real actor or influencer? No — models block real people and celebrities. Build original characters instead.

One long multi-shot clip or many separate clips? Single-pass holds identity internally but is length-capped; separate generations need the lock block plus identical references.

Do first/last frame controls help? Yes — pinned first/last frames (Seedance 2.5, Hailuo 3) anchor how shots start and end.

Build the character once, reuse the lock block forever. FLOWNIB can push finished episodes across platforms in one pass.