Plimo

Grok Imagine: Make Grok AI Videos with Sound on Grok Imagine 1.5, Free to Start

Two Friends Hug
Bring Photo to Life
Paper Boat in the Rain
Yearbook
Muscle
Hand Heart
Bald Transition
Ice Queen
Billboard
Cinematic Push-in
Bullet Time
Kung Fu
Gold Mermaid
Exclusive First Class
Older Twin
Red Carpet Moment

What is Grok Imagine?

Grok Imagine is xAI's family of video and image models, the same name you see on the Imagine feature in the Grok app. Its video side is known for lively motion and for putting sound into every clip, so a scene arrives with voices, footsteps or music already in it. On Plimo you get Grok Imagine 1.5 for video, plus Image 2.0 and the lighter Grok Imagine Image for pictures. The generator above opens in video mode on version 1.5.

  • Grok Imagine 1.5

    Lively motion with sound.

    • 6–15 s
    • 480P / 720P
    • Always with sound
    • Up to 7 reference photos

    From 60 credits

    Free credits work: ⁨480P⁩ · ⁨6 s⁩
  • A corgi on a surfboard riding a small wave at a sunny beach

    Grok Imagine Image 2.0

    Vivid pictures, and edits from up to 3 photos.

    • 1K / 2K
    • Up to 3 reference photos

    From 20 credits

  • A corgi on a surfboard riding a small wave at a sunny beach

    Grok Imagine Image

    Simple, affordable images, and edits from up to 5 photos.

    • 1K
    • Up to 5 reference photos

    From 9 credits

How xAI's Imagine models evolved

xAI has shipped a new release every few months since late 2024.

  1. December 2024

    Aurora

    xAI's first in-house image model inside Grok, built for photorealistic scenes and close prompt following, and able to edit pictures people uploaded.

  2. August 2025

    Imagine in the Grok app

    Image and video generation arrived under the Imagine name, turning pictures into short clips that came with sound.

  3. October 2025

    Version 0.9

    A big step up in picture quality and motion, with synchronized audio that could include dialogue and singing.

  4. February 2026

    Grok Imagine 1.0

    Clips grew to 10 seconds with a sharper picture, voices became more expressive, background music fit the scene better, and long prompts were followed more closely.

  5. June 2026

    Grok Imagine 1.5

    After a short preview, xAI released it with clips of up to 15 seconds, stronger image-to-video results and native audio in every clip.

  6. August 2026

    Image 2.0

    The second-generation picture model, aimed at cleaner, readable text inside images and at more precise editing from your own photos.

Making a Grok video on Plimo

  1. Describe the scene and its sound

    Version 1.5 is already picked in the generator at the top. Write what happens and what we should hear, like rain drumming on a tin roof or a crowd counting down to midnight.

  2. Add reference photos if you have them

    Sign in and drop up to 7 photos into the slot beside the prompt box (above it on a phone). Once photos are attached, the clip takes its shape from them instead of the aspect ratio setting.

  3. Choose length and resolution

    Anything from 6 to 15 seconds works, at either of two resolutions, and the credit count on the Generate button follows your choices.

  4. Flip to Image for still pictures

    Switch the generator from Video to Image and it moves to Image 2.0. The model menu also holds the basic Image model and every other model on Plimo.

  5. Collect the results

    Finished clips and pictures land in My creations below. Download the keepers, or press Remake to reload the prompt and settings.

Why use Grok on Plimo

  • Your first clip is on us

    New accounts receive 100 free credits once, which is enough for one 6-second clip on version 1.5 at 480p. Signing up asks for no card.

  • Video and stills share one account

    The video model and both picture models share one balance and one creations list, so a picture made here can be the reference photo for your next clip.

  • Try another model on the same idea

    If Grok's take isn't right, run the prompt on Seedance, Kling or Veo from the same menu and keep the one you like.

  • The cost shows up front

    The Generate button lists the credits for this exact clip or picture before you press it. A failed run hands the credits back.

  • No watermark on member work

    Creations you make as a member download without a watermark and can be used commercially. Work made without a membership carries a small mark and is for personal use.

How Grok compares with other video models

Each has its own page; in short, here is how they differ.

  • Kling

    Kling 4.0 runs from 3 to 30 seconds and can copy the movement in a clip you upload, while Grok tops out at 15 seconds and works from text or photos only.

  • Veo

    Veo 3.1 Fast always delivers 8 seconds with a polished, cinematic look and can land on an end frame you choose; Grok gives you a range of lengths instead.

  • Seedance

    Seedance 2.0 Mini is the other video model new accounts can try on free credits, and Seedance takes reference clips as well as photos.

  • Hailuo AI

    Hailuo AI's MiniMax H3 Max leans toward realistic faces and mixes reference photos with short clips, where Grok is more about energetic movement.

  • Wan

    Wan 3.0 reaches 30 seconds, takes start and end frames plus reference clips, and lets you switch the audio off, which Grok never does.

  • Vidu

    Vidu Q4 Preview accepts up to 15 reference photos for a clip of up to 16 seconds, where Grok takes 7.

  • FLUX

    FLUX 3 Video works from text or a start and end frame for 5 to 20 seconds, with sound you can turn off.

  • SkyReels

    SkyReels V4 also puts sound in every clip and offers a start-and-end-frame mode as well as a reference mode with up to 3 photos.

  • HappyHorse

    HappyHorse 1.1 needs a written prompt even when you give it a photo, and blends up to 9 reference photos into one clip with sound.

What people make with Grok

  • A pet with something to say

    Add a photo of your dog and describe it turning to the camera and grumbling about bath time. The voice and the splashing come built into the clip.

  • A street scene you can hear

    Describe a night market in the rain with sizzling grills, chatter and a scooter rolling past. From text alone, the 1.5 model builds the picture and the noise around it.

  • A poster that comes to life

    Make a stormy lighthouse poster with Image 2.0, switch back to Video, attach it as a reference photo and ask for waves crashing while the beam sweeps across the sea.

  • A birthday shout-out

    Upload a photo of your friend and describe confetti drifting down while a small crowd sings happy birthday off-screen. Six seconds is plenty for a story post.

  • Signs and menus with real words

    Image 2.0 copes well with short text, so ask for a café chalkboard that reads Soup of the Day or a gig flyer with your band's name.

  • Lots of quick concept pictures

    When you want many rough options, the basic Image model is the cheaper one. Try looks for a bedroom, a character or a logo idea, then redo the winner in Image 2.0.

Common questions about Grok

No. Plimo is an independent website that connects to xAI's models through an API. It isn't made by xAI and has no affiliation with it.

Partly. The 100 free credits every new account gets once cover one Grok Imagine 1.5 video that is 6 seconds long at 480p. Longer or sharper clips and both picture models need a membership, explained on pricing. Nano Banana 2 is a picture model that also works on free credits.

Each version card above shows a starting price, and the Generate button shows the exact amount before you submit. Longer or sharper clips cost more; for pictures, size and added photos change the price.

Yes. The Grok AI video generator here runs version 1.5, which makes clips from 6 to 15 seconds, set in whole seconds under Length, at 480p or 720p.

Yes. There is no audio switch, so each clip comes with sound. Mention what you want to hear, such as a spoken line, music or background noise, and the model tries to match it.

In video mode, add up to 7 photos as references and say how the scene should move; leave the slot empty and a prompt alone is enough. With photos attached, the clip follows their proportions rather than the aspect ratio you picked. More models that animate a picture are on image to video.

Both make this page a Grok AI image generator, in different ways. Image 2.0 is newer, with a larger size option and stronger detail and lettering, and it can edit using up to 3 of your photos. The basic Image model makes one size only, accepts up to 5 photos and costs less per picture. For more picture models, open the AI image generator.

Without a membership, videos and pictures carry a small mark in the bottom-right corner and are for personal use only. What you create while you're a member comes as clean files you can use commercially.

Plimo doesn't allow adult content, on Grok or any other model. If the model turns down a prompt or photo after you submit, the credits go back to your balance.

More tools and models for video and images

All models