Spoken lines, in sync
Lines in double quotes come out spoken with lip sync, and two people can trade lines in one clip. In our tests every line was said word for word.
Grok Imagine · by xAI
Grok Imagine AI video generator
Grok Imagine is xAI's video and image model. Its video side is known for short, phone-style clips where people speak, sing and react with the sound made in the same pass, at a low price per second. Write a scene below and it opens in Video Studio on Grok Imagine Video 1.5.
Example video
Opens in Video Studio with the model you picked. You see the price in credits before anything runs.
What Grok Imagine is known for
Sound is always on
Voices, room tone, music and effects come with every clip. There is nothing to switch on and no extra charge for it.
Cheap enough to take many shots
Video 1.5 costs a fraction of the flagship models per second, and 1.5 Lite costs less again. It suits volume: hooks, reactions, social posts.
Phone-video realism
Handheld vlogs, selfie clips and candid moments look like something filmed on a phone, which is what most social video is.
Where it's weaker
- It sometimes burns subtitles into the picture when someone speaks, more often in cinematic scenes. Write "no subtitles or on-screen text"; if they still appear, run it again.
- Weaker than Seedance and Kling on physics, effects and multi-shot continuity; reviewers find it hard to recommend for film work.
- Faces and hands can blur in fast motion and close-ups.
- Mirror shots and reflections break easily.
- Clips stop at 15 seconds.
- xAI's consumer app has faced investigations over sexual deepfakes; on Affogato the safety filter is always on.
- Maker
- xAI · Video 1.5 released June 2026
- Clip lengths on Affogato
- 5, 6, 8, 10 or 15 s
- Aspect ratios
- 9:16 · 2:3 · 3:4 · 1:1 · 4:3 · 3:2 · 16:9
- Sound
- Always on: dialogue with lip sync, effects, music
- Inputs
- Text · a start image · up to 7 reference images
- Image models
- Grok Imagine Image 2.0 and Image, in Image Studio
Made with Grok Imagine on Affogato
Every clip below was generated with the prompt shown, unedited. Load any prompt into the generator and make your own version.
Two people talking
Saturday at the barbershop
Prompt
Vertical phone video in a busy neighborhood barbershop. A barber in his fifties with a grey goatee trims a young customer's fade and says to the camera, "Every Saturday this kid tells me he wants it shorter." The customer grins in the mirror and says, "And every Saturday you don't listen." The barber laughs and points the clippers at him. Warm fluorescent light, posters on the wall. Sound: clippers buzzing, their voices, a radio playing old soul.
Food-vlog reaction
First bite at the night market
Prompt
Vertical handheld food-vlog clip at a night market in Bangkok. A young woman with a nose ring and a tote bag takes a bite of grilled squid on a stick, her eyes go wide, and she says to the camera, "Oh no. Oh no, that's so good." She laughs with her mouth half full and gives a thumbs up. Neon signs and smoke behind her. Sound: sizzling grills, crowd chatter, her voice.
Singing
Porch blues with grandpa
Prompt
Vertical clip on a sunny porch: a grandfather in his eighties with a white mustache plays a slow blues riff on an old acoustic guitar while his granddaughter, about eight, sits beside him and taps her foot. He sings softly, "Well the sun comes up, and the coffee's on." She giggles and tries to sing along. Warm late-afternoon light, the wooden porch creaks. Sound: acoustic guitar, his gravelly voice, her laugh.
Selfie vlog
Five dogs, one walker
Prompt
Vertical phone vlog in a city park in autumn. A woman in her twenties in a yellow raincoat walks five dogs of different sizes at once, the leashes tangling. She turns the phone to herself, laughing, and says, "This is my cardio now." A tiny terrier jumps up at her knee. Overcast light, wet leaves. Sound: her laugh, jingling collars, a dog barking. No subtitles or on-screen text.
A line in a busy room
"Salt. Then it's perfect."
Prompt
Vertical phone video in a busy restaurant kitchen. A head chef in his forties with tattooed forearms tastes a sauce from a spoon, pauses, looks at the young cook beside him and says, "Salt. Then it's perfect." The cook nods fast and reaches for the salt. Steam, stainless steel, warm light. Sound: sizzling pans, the chef's low voice, the clatter of a busy kitchen. No subtitles or on-screen text.
Action
Ollie at sunset
Prompt
Vertical slow-motion clip: a teenage skateboarder with long braids and a baggy flannel shirt rolls toward the camera in an empty skatepark at sunset, pops an ollie over a small gap, lands it and throws both arms up, laughing. Golden backlight, dust in the air. Sound: wheels on concrete, the clack of the board, her shout.
Dance trend
Living-room dance, one missed move
Prompt
Vertical phone video in a sunny living room: two teenage sisters in matching oversized hoodies do a quick synchronized dance routine, then one messes up the last move and they both collapse laughing on the sofa. The phone is propped on a shelf. Natural daylight, family photos on the wall. Sound: an upbeat pop beat from a speaker, their laughter. No subtitles or on-screen text.
Which Grok Imagine to pick
Start here
Grok Imagine Video 1.5
Most clips
The current model: talking, reacting and singing people with sound, from text, a photo or references.
5 s clip 127 credits · 5–15 s
Open with Grok Imagine Video 1.5Option 2
Grok Imagine Video 1.5 Lite
Drafts and volume
A lighter, much cheaper version of 1.5. Try hooks and ideas here, then re-run the keepers on 1.5.
5 s clip 28 credits · 5–15 s
Open with Grok Imagine Video 1.5 LiteOption 3
xAI Grok Imagine Video
Extending and editing clips
The first Grok Imagine video model. Still the one to use for extending a clip.
5 s clip 64 credits · 5–10 s
Open with xAI Grok Imagine VideoGrok Imagine Image 2.0 and the original Grok Imagine Image, with their edit versions, are in Image Studio's model menu. Credits are read from the live catalog; the Standard plan ($24/month) includes 2,000 credits, about 15 five-second Grok Imagine Video 1.5 clips.
Prompting Grok Imagine well
- Write it like a phone video: who holds the camera, who is in frame, then what they say and do.
- Put each spoken line in double quotes and keep it short; one or two lines per person fits the length.
- Add "no subtitles or on-screen text" to every prompt with dialogue.
- Name the sounds you want (clippers buzzing, a radio, crowd chatter); it builds the room from them.
- Prefer handheld, selfie and candid setups; it is weaker on big cinematic camera moves and effects.
- Draft on 1.5 Lite, then run the keeper on 1.5.
- Start from a photo when a face, outfit or product has to stay the same.
From prompt to clip
Describe the scene or add a photo
Write the clip in the box at the top of this page, or start from an example. Add a photo to use it as the first frame.
Sign in and land in Video Studio
Your prompt, photo, format and the Grok version you picked carry over. Nothing runs until you press Generate.
Pick the length, then generate
Choose a length up to 15 seconds. The credit price shows under the Generate button before anything is spent.
Keep the best take
Clips are cheap enough to run a few takes. Keep the one where the line lands, then extend it or add it to a longer edit.
Questions about Grok Imagine
What is Grok Imagine best at?
Short, phone-style clips of people talking, reacting and singing, with lip-synced dialogue and sound in one pass, at a low price per second. It is weaker on physics, effects and long cinematic sequences.
What is the difference between Grok Imagine Video 1.5 and 1.5 Lite?
Lite is a lighter version of 1.5 that costs much less per second. Use it for drafts and volume, and 1.5 when the take matters.
Why does my clip have subtitles burned in?
Grok Imagine sometimes adds captions when someone speaks. Write "no subtitles or on-screen text" in the prompt and run it again; it usually goes away.
Is Grok Imagine better than Seedance or Kling?
Not for cinematic work: reviewers and our own tests favour Seedance 2.0 and Kling 3.0 for physics, action and multi-shot scenes. Grok Imagine is the cheaper pick for talking-head and social clips.
Does Grok Imagine make images too?
Yes. Grok Imagine Image 2.0 and the original Grok Imagine Image, with their edit versions, are in Image Studio.
Is Grok Imagine free on Affogato?
No. Every run costs credits, priced by the second and shown before you generate. Credits come with a monthly plan and unused credits roll over up to the plan's cap.
Can I use Grok Imagine videos commercially?
Yes. Paid Affogato plans include a commercial license and no watermark, within our terms of service and xAI's content rules.
Keep exploring
Facts checked November 15, 2026. Sources: fal: Grok Imagine Video 1.5 · fal: how to use Grok Imagine · fal: Grok Imagine launch on fal · GIGAZINE: Grok Imagine Video 1.5 · GIGAZINE: Grok Imagine 1.0 · Curious Refuge: Grok Imagine 1.5 review · The Register: Ofcom opens Grok probe