Skip to content

Early Black Friday: +100 bonus credits on the First Film Pack.Early Black Friday A one-time Black Friday bonus, one per account, at the standard US$ price.

Get +100 bonus credits
Tutorials
Model Guide · Grok Imagine Video 1.5

How to prompt Grok Imagine Video 1.5.

FXI StudioSeptember 19, 2026

Grok Imagine Video 1.5 renders realistic motion with synchronized audio in a single pass, holds tight to your prompt, and handles complex movement other models flatten. That precision is a gift and a discipline: it does exactly what you write, so the craft is in writing it well. Here is how we direct it inside FXI Studio.

The short answer

Grok Imagine Video 1.5 is an image-to-video AI model that renders realistic motion with synchronized audio in one pass. To prompt it, start with a still first frame, then direct what happens next: keep it to one subject and one camera move, use intensity modifiers to dial the performance, name the camera move, and write the Sound line like a sound designer.

Burdock · Rendered in FXI Studio · Grok 1.5

Burdock, the closer. One subject, one restrained gesture, one push-in — rendered end to end with synchronized sound. The prompt is below.

Try this prompt

“A young chef plates her first dish in a busy restaurant kitchen, pauses, and breaks into a proud smile as the head chef nods. Slow push-in on a 50mm lens, warm tungsten key light. Sound: sizzling pans, a soft exhale, the bustle of a busy kitchen.”

Make this — free

Opens the studio with this prompt filled in · 65 free credits

Video examples

Seven prompts, one rendered, six to steal.

The first example is the Burdock clip above, fully rendered. The rest are recipes: the exact prompt, paired with the character reference image you would feed as the first frame. Copy, adapt, remix.

Rendered · Grok 1.5
Example 01 · Rendered

The closer

A power beat, held on a single face.

This is the marquee clip, rendered end to end in FXI Studio. One subject, one restrained gesture, one camera move. Grok holds the face, syncs the breath to the motion, and lets the room tone do the tension.

Burdock, a dark-complexion bearded executive in a charcoal three-piece suit, slowly unbuttons his jacket and squares his shoulders toward camera, chin lowering a fraction as a faint knowing smile crosses his scarred brow. Slow push-in on a 50mm lens, shallow depth of field. Low warm tungsten key from frame left, deep shadow pooling behind him in a wood-paneled boardroom at dusk. Sound: the soft creak of leather, one measured exhale, a distant grandfather clock ticking, a low room tone that tightens as the camera moves in.
High noon — character reference
First frame
Jax Ryder — character reference. Feed this as the first frame.
Example 02 · Recipe

High noon

Stillness that reads as a threat.

A pure tension recipe. The intensity lives in what does not move — Grok reads “dead still,” “hovering,” and “narrowing hard” as micro-motion, not stasis. Save the whip-pan for the release.

Jax Ryder, a rugged charismatic Black gunslinger in a dust-worn duster, stands dead still at the end of a sun-bleached frontier street, hand hovering an inch above his holster, eyes narrowing hard against the glare. Locked-off wide, then a fast whip-pan to his eyeline. Blistering high-noon sun, heat haze warping the far storefronts, dust drifting low across the frame. Sound: a lone wind gust, a saloon door creaking on its hinge, spurs shifting once in the dirt, no music until the very last beat.
The editorial turn — character reference
First frame
Charifa — character reference. Feed this as the first frame.
Example 03 · Recipe

The editorial turn

Fashion motion, controlled to the millimeter.

Editorial motion is about restraint. “Turns smoothly,” “one slow blink,” “holds the gaze” give Grok a clean arc with a clear end state, so the face lands instead of drifting.

Charifa, a high-fashion Black woman with a sharp dark bob and blunt bangs, hazel-green eyes, in a black leather high-neck, turns her head smoothly toward camera and holds the gaze, one slow blink, jaw lifting with quiet confidence. Locked 85mm portrait lens, gentle rack focus from shoulder to eyes. Cool key from a large softbox camera-right, one hard rim light carving her profile against seamless charcoal. Sound: the faint hush of a studio, the soft click of a shutter on the hold, a low cinematic swell arriving under the final frame.
The speakeasy solo — character reference
First frame
Parsley — character reference. Feed this as the first frame.
Example 04 · Recipe

The speakeasy solo

Where the Sound line does the acting.

Grok renders synchronized audio in the same pass, so a musician is the ideal stress test. Name the instrument, the phrasing, and the room. The trumpet locks to the lip and the shoulder roll.

Parsley, a late-20s New Orleans jazz trumpeter in a navy suit and gold round glasses with a handlebar mustache, lifts his trumpet and leans into a soaring solo, shoulders rolling with the phrase, eyes closing at the peak. Slow orbit around him from the side, drifting from mid-shot to close. Smoky amber speakeasy haze, a single warm spotlight, motes drifting through the beam. Sound: a live muted trumpet climbing a bluesy phrase, brushed snare and upright bass underneath, low crowd murmur, a single glass clink two bars in.
The porch, at dusk — character reference
First frame
Pop Hollis — character reference. Feed this as the first frame.
Example 05 · Recipe

The porch, at dusk

A quiet human moment, fully scored by ambience.

Warmth is a difficult register for AI video, which tends to overplay it. Under-describe the emotion (“a warm, crinkling smile”) and over-describe the ambience. The crickets and the chair creak carry the scene.

Pop Hollis, a 70-year-old Midwestern grandfather with silver hair in a worn flannel, a weathered kind face, rocks slowly in a wooden porch chair and breaks into a warm, crinkling smile as he looks off toward the fields. Static wide, then a gentle slow push to a medium. Golden-hour side light, long shadows across the porch boards, dust catching the last sun. Sound: crickets starting up, a screen door tapping in the breeze, the creak of the rocking chair, a distant dog bark, no music at all.
Salt wind — character reference
First frame
Kalei — character reference. Feed this as the first frame.
Example 06 · Recipe

Salt wind

Natural motion, cued by weather.

Weather is free motion. “A gust lifts her hair,” “salt spray hanging” give Grok physical forces to animate, which reads far more alive than asking a character to simply “move.”

Kalei, a mid-20s Hawaiian woman with long dark curls and sun-kissed skin, laughs freely as an offshore gust lifts her hair across her face, one hand rising to sweep it back, weight shifting toward the horizon. Handheld medium, a slow drift left following her turn. Blown-out golden coastal light, salt spray hanging in the air, lens flare skating across frame. Sound: wind buffeting the mic in bursts, waves breaking behind her, gulls calling once, her light unguarded laugh over the top.
The verdict — character reference
First frame
Taro — character reference. Feed this as the first frame.
Example 07 · Recipe

The verdict

Dialogue, lip-synced in one take.

Grok lip-syncs spoken lines. Write the dialogue in quotes exactly as it should land, keep it to one or two sentences, and describe the delivery (“calmly and deliberately”) so the performance matches the words.

Taro, a dignified late-60s man with a salt-and-pepper beard in a gray suit, leans forward at a heavy desk in a wood-paneled study and says, calmly and deliberately, "The decision is already made. I only asked you here to see whether you knew it." He holds the look after the line, one eyebrow rising a fraction. Slow push-in on an 85mm lens. Low warm lamplight from a green banker’s lamp, deep amber shadow behind. Sound: his measured baritone voice, the faint hum of the lamp, rain against a far window, a clock ticking under the silence.
How to prompt it

Five habits that make Grok behave.

01

Start with the image

Grok Imagine Video 1.5 is strongest as an image-to-video model. Give it a still first frame — a character portrait, an establishing plate, a rendered still — and write the prompt as direction for what happens next, not a description of what already exists. Every character above ships with a reference image for exactly this: feed the still, then describe the move. The model preserves the face, the wardrobe, and the lighting from the frame, so you spend your words on motion and sound instead of re-describing a person the image already shows.

02

Keep it focused

One subject, one action, one camera move per clip. Grok adheres tightly to what you ask, which cuts both ways: ask for three things happening at once and it will attempt all three and land none. A closer unbuttons a jacket. A trumpeter leans into a solo. A grandfather breaks into a smile. Resist the urge to choreograph a whole scene into a single generation. Chain shorter, cleaner beats instead, then stitch them in the edit.

03

Use intensity modifiers

The difference between “he turns his head” and a directed performance is the adverb. Slowly, sharply, a fraction, freely, deliberately, dead still. These modifiers are how you dial a movement up or down, and Grok reads them faithfully. “Chin lowering a fraction” is a completely different beat than “snapping his head down.” Direct the intensity the way you would direct an actor: not just what moves, but how much, and how fast.

Vague

“She turns and smiles.”

Specific

“She turns her head smoothly toward camera and holds the gaze, one slow blink, the smile arriving late and small.”

04

Describe camera movement

Name the shot and name the move. Slow push-in, locked-off wide, gentle orbit, whip-pan, handheld drift. Grok treats the camera as a first-class element, so an unstated camera defaults to something generic and floaty. Pair the move with a lens where it matters — a 50mm push-in reads differently than an 85mm one — and give the motion a reason to travel: a face to land on, a horizon to drift toward, an eyeline to snap to. One camera grammar per clip, never two.

Vague

“A man stands in a boardroom.”

Specific

“Slow push-in on a 50mm lens, shallow depth of field, settling on his face as the room tone tightens.”

05

Write the Sound line like a sound designer

Grok renders synchronized audio in the same pass, and the Sound line is where most clips are won or lost. Do not write “ambient noise.” Layer it the way a mixer would: the key sound (a voice, a trumpet, a footstep), the bed (room tone, wind, rain), and one specific accent (a glass clink, a door creak, a single gull). Then decide what the music does, or whether there is any at all. A well-built Sound line turns a moving image into a scene.

Vague

“Sound: background music and city noise.”

Specific

“Sound: a live muted trumpet climbing a bluesy phrase, brushed snare and upright bass underneath, low crowd murmur, a single glass clink two bars in.”

Frequently Asked

Grok Imagine Video 1.5, answered.

What is Grok Imagine Video 1.5?

Grok Imagine Video 1.5 is an AI video model that turns a still image and a text prompt into realistic motion with synchronized audio in a single pass. It is strongest as an image-to-video model: you feed a first frame, then describe the movement and sound you want.

How do you prompt Grok Imagine Video 1.5?

Prompt Grok Imagine Video 1.5 by starting with a still first frame, then writing direction for what happens next. Keep it to one subject, one action, and one camera move; use intensity modifiers like "slowly" or "a fraction" to dial performance; name the camera move; and write a detailed Sound line with a key sound, a bed, and one accent.

Does Grok Imagine Video 1.5 generate sound?

Yes. Grok Imagine Video 1.5 renders synchronized audio in the same pass as the video, including lip-synced dialogue. Write the spoken line in quotes exactly as it should land and describe the delivery so the performance matches the words.

Is Grok Imagine Video 1.5 image-to-video or text-to-video?

Grok Imagine Video 1.5 works best as an image-to-video model. Give it a still first frame — a character portrait or an establishing plate — and it preserves the face, wardrobe, and lighting while animating the motion you describe.

How do I try Grok Imagine Video 1.5?

You can try Grok Imagine Video 1.5 in FXI Studio. New accounts start with 65 free credits, and paid plans run $15, $35, and $135 per month. Every prompt in this guide is reproducible inside the studio.

Want the six canonical camera moves and their prompt structures? Read the companion Seedance 2.0 shot-grammar guide, or open the studio and check pricing.

Now go direct one

Give it a still, a direction, and a Sound line.

Every prompt on this page is reproducible inside FXI Studio. Start with 65 free credits, feed a character reference, and write the move you already see. Plans run $15, $35, and $135 a month when you scale.

Make this — freeStart free — 65 credits

Related guides

Looking for a specific tool? Browse the documentation, the guides, or all tutorials.