Up to 30 seconds per video
Set the length to the second, from 6 to 30 seconds (10 s by default): enough for an ambient loop, a full scene or a welcome video.
Video model · xAI
Grok Video — 6 to 30-second videos with sound, from text or a photo.
Grok Video is Deepnia’s name for xAI’s Grok Imagine video model, opened to developers in January 2026. On Deepnia it creates videos of 6 to 30 seconds with sound, set to the second, from text or a photo to animate, in 480p or 720p and 5 aspect ratios, including 2:3 portrait and 1:1 square.
- 6–30 s
- 480p or 720p
- 5 ratios incl. 2:3
- Text or 1 photo
- Sound included
- Length set to the second
The video tool opens with Grok Video already selected. The exact cost shows before every creation; you sign up when you generate.
- A text prompt or a photo
- 6–30 s, 5 aspect ratios
- A video in 480p or 720p
At a glance
- Maker
- xAI
- Length
- 6 to 30 s (10 s by default)
- Resolution
- 480p or 720p (480p by default)
- Aspect ratios
- 9:16, 1:1, 16:9, 2:3, 3:2 (2:3 by default)
- Modes
- Normal, Fun, Spicy (Normal by default)
- Inputs
- Text, or 1 photo to animate
- Sound
- Included in every mode
What Grok Video does best
Sound in every video
Every video comes with its own sound, whatever the mode: speech, music, sound effects and ambience. Describe the sounds and lines you want to hear in the prompt.
Bring a photo to life
Add a photo: xAI says it becomes the first frame of the video. On Deepnia the video then keeps the photo’s aspect ratio, so crop it before adding it (up to 10 MB).
A format for every feed
With text only, 5 aspect ratios: 9:16, 1:1, 16:9, 2:3, 3:2. The 2:3 portrait, selected by default, suits news feeds, 9:16 suits stories and Reels, and 16:9 suits YouTube.
Built for fast iteration
xAI says it tuned Grok Imagine for speed and request volume as well as quality. Handy for running drafts at 480p before the final cut at 720p.
Camera moves written in the prompt
xAI’s examples show zooms, dolly shots, tilts, pans and timelapses, simply described in the prompt.
How to use Grok Video on Deepnia
Open the video tool
The “Try Grok Video” button opens Deepnia’s video tool with the model already selected.
Text only or a photo
Write your prompt, or add a photo to animate (up to 10 MB): the video then keeps its aspect ratio.
Choose the mode
Under “Mode”, keep Normal, the default, or switch to Fun.
Set it up and generate
Set the length (6–30 s), resolution and aspect ratio, check the cost shown, then create.
Grok Video on Deepnia: what you can set
- Inputs
- Text only, or text plus 1 photo to animate
- Length
- 6 to 30 s, to the second (10 s by default)
- Resolution
- 480p or 720p (480p by default)
- Aspect ratios
- 9:16, 1:1, 16:9, 2:3, 3:2 (2:3 by default); with a photo, the video keeps its aspect ratio
- Modes
- Normal, Fun, Spicy (Normal by default); Spicy mode starts from text only
- Photo
- 1 at most, up to 10 MB; crop it before adding it
- Sound
- Included in every mode: speech, music, sound effects and ambience
- Prompt
- Required, up to 5,000 characters
Prompt ideas to try
Copy a prompt, open the tool and adapt it to your project.
Morning mood on a shopping street
- Normal
- 16:9
- 20 s
- 720p
- Text only
A shopping street early in the morning: shopkeepers open their stalls of fabric and fruit, bicycles pass by, golden light shines through the dust. The camera moves slowly down the middle of the street, realistic documentary style.
Animated product photo for a leather brand
- Normal
- Photo’s aspect ratio
- 8 s
- 720p
- 1 photo of the bag
Animate this photo: the woven leather bag slowly turns on itself, light glides over the stitching and the shine of the leather, slight zoom in at the end.
Comedy sketch for social media
- Fun
- 9:16
- 10 s
- 480p
- Text only
A very serious goat in a suit and tie presents the TV news in a small studio, exaggerated gestures, glasses sliding down its snout, comic mood.
Portrait catwalk for a fashion label
- Normal
- 2:3
- 10 s
- 720p
- Text only
Full-length portrait shot: a woman walks toward the camera down a hallway lined with colorful fabrics, her long dress floating with every step, the camera slowly moving back ahead of her.
Welcome loop for an evening in Cape Town
- Normal
- 16:9
- 30 s
- 720p
- Text only
A lantern-lit terrace facing the ocean at sunset in Cape Town: guests arrive in evening wear, waiters pass with trays, the camera glides slowly and continuously along the terrace.
Ice cream shop mascot, quick test
- Fun
- 9:16
- 6 s
- 480p
- Text only
An ice cream cone mascot dances in front of a pastel-colored shop, exaggerated moves, flying confetti, the camera circling around it, 3D cartoon style.
Tips for better results
- Describe the camera move in the prompt (zoom, dolly, tilt, pan): that is how xAI directs its own examples.
- Describe the sound in the prompt too: the ambience, the music, the sound effects or a character’s line in quotation marks.
- Test the idea at 480p and 6 s, then run it again at 720p with the final length.
- With a photo, focus on what happens next: according to xAI, the photo serves as the first frame.
- Crop the photo to the aspect ratio you want before adding it: on Deepnia, the video keeps the image’s aspect ratio.
- For a long video, describe one continuous action (a tracking shot, guests arriving) rather than a string of unrelated scenes.
Good to know
- To tell a story longer than 30 seconds, create several videos and join them in your editor.
- For a series of videos with the same character or product, start from the same photo each time.
- Spicy mode starts from text only; with a photo, choose Normal or Fun.
- Use photos you have the rights to, especially when animating a real person.
Grok Video or another model?
Deepnia’s other video models, to pick the right one for your project.
| Model | Length | Resolution | Aspect ratios | Best for |
|---|---|---|---|---|
| Grok VideoThis model | 6–30 s | 480p–720p | 9:16 · 1:1 · 16:9 · 2:3 · 3:2 | 6 to 30-second videos with sound, from text or a photo. |
| Seedance 2.5 | 4–30 s | 480p–720p | 16:9 · 9:16 · 1:1 · 3:4 · 4:3 · 21:9 | Also up to 30 s, guided by up to 16 image and video references. |
| Gemini Omni | 4, 6, 8 or 10 s | 720p–4K | 9:16 · 16:9 | Short clips up to 4K, guided by 7 reference images. |
| Veo | 8 s | Set by the mode | 9:16 · 16:9 | 8 s clips with sound, in three quality levels. |
| Kling 3.0 | 3–15 s | Set by the mode | 9:16 · 1:1 · 16:9 | A 5-shot editor, persistent characters and optional sound. |
| HappyHorse | 3–15 s | 720p–1080p | 9:16 · 1:1 · 16:9 · 4:3 · 3:4 | Editing an existing 3–60 s video. |
Frequently asked questions
What is Grok Video?
It is Deepnia’s name for xAI’s Grok Imagine video model, which creates a video from text or a photo. xAI opened Grok Imagine to developers in January 2026.
How long can a Grok Video clip be?
From 6 to 30 seconds, set to the second, 10 seconds by default.
What resolution do the videos come in?
480p (the default) or 720p. Test the idea at 480p, then run it again at 720p for the final cut.
Can I animate a photo with Grok?
Yes, with a photo of up to 10 MB. According to xAI, it becomes the first frame of the video; on Deepnia the video keeps its aspect ratio, so crop it before adding it.
Which modes does Grok Video offer?
Grok Video offers 3: Normal, Fun, Spicy, with Normal by default, and each one gives you a video with sound. Try the same prompt in Normal and then in Fun on a short test to pick the result you like. Spicy mode starts from text only.
Which aspect ratios are available?
With text only: 9:16, 1:1, 16:9, 2:3, 3:2, with 2:3 selected by default. With a photo, the video keeps the image’s aspect ratio.
Do Grok videos have sound?
Yes. Every video is generated with its own sound, in every mode: speech, music, sound effects and ambience, with nothing to switch on. Describe the sounds and lines you want to hear in the prompt.
How do I keep the same character across several videos?
Start from the same photo each time: according to xAI, it serves as the first frame. Describe the character with the same words from one prompt to the next, too.
Ready to try Grok Video?
The video tool opens with Grok Video already selected. The exact cost shows before every creation; you sign up when you generate.