1
Choose Text-to-Video or Image-to-Video
Pick text-to-video to start from a prompt, or image-to-video to animate up to seven reference images with the Grok video generator.
Grok ImagineGrok ImagineTurn a written scene or reference image into a short video
Create Grok videos from a text prompt or animate reference images with image-to-video. Direct the action, camera movement, mood, and sound in your prompt, then choose 6–30 seconds, 480p or 720p, and Normal, Fun, or Spicy mode.
Create Grok AI Videos in 4 Simple Steps
Pick text-to-video to start from a prompt, or image-to-video to animate up to seven reference images with the Grok video generator.
Describe the scene, action, camera movement, mood, and the sounds you want — Grok generates audio from your description too.
Choose 6–30 seconds, 480p or 720p, and Normal, Fun, or Spicy mode. Text-to-video also lets you choose one of five aspect ratios.
Preview the complete clip with sound, check motion and small visual details, then download it for editing, publishing, or another stage in your workflow.
Prompted Motion, Reference Images, Native Audio, and Flexible Formats
Describe the subject, action, setting, camera behavior, lighting, and sound. The generator uses that prompt to create a new short video from scratch.
Upload up to seven reference images to guide the subject, look, or composition, then describe how the scene and camera should move. Review identity and product details in the final clip.
Describe dialogue, ambience, sound effects, or musical mood alongside the visual direction. Grok can generate audio with the video, although important speech and timing should be reviewed.
Use plain language for pans, tilts, tracking shots, dolly moves, focus changes, subject motion, and scene transitions. Clear motion cues help avoid a static-looking result.
Choose Normal, Fun, or Spicy in the generator. Treat each as a creative direction, then adjust your prompt and review the result against the platform's content rules.
Generate 6–30 second clips at 480p or 720p. Text-to-video supports 16:9, 9:16, 1:1, 3:2, and 2:3 for landscape, vertical, square, and portrait layouts.
How to Use Grok to Create AI Videos
The Grok video generator supports both text-to-video and image-to-video in one browser workflow. Use these steps to direct the visuals and audio, select the output settings, and review the completed clip.
Open the playground at the top of this page. Nothing needs to be installed; sign in when you are ready to generate your first Grok video.
Start from a text prompt for a new scene, or switch to image-to-video and upload up to seven images to guide the subject, style, framing, or product appearance.
Write the action, camera movement, mood, and visual style. Also describe any dialogue, ambient noise, sound effects, or musical mood you want the model to attempt.
Choose 6–30 seconds, 480p or 720p, and Normal, Fun, or Spicy mode. In text-to-video, choose 9:16 for vertical video, 16:9 for landscape, or another supported ratio.
Submit the generation, preview the finished video with sound, and check faces, hands, products, text, and timing. Refine the prompt if needed, then download the clip.
Tell the model what moves and how the camera behaves — 'slow dolly-in', 'handheld pan', 'subject turns toward camera'. Motion cues produce far more dynamic clips than static descriptions.
When a face, product, or look matters, start from clear reference images. They provide visual guidance, but you should still inspect consistency across the entire clip.
Spell out the audio — 'rain on a window, distant thunder', 'upbeat synth music', or a short line of dialogue. Keep sound direction distinct from the visual instructions.
Begin at 6 seconds to test the look and motion, then generate a longer version once the prompt is working. Short tests make it easier to refine one change at a time.
Grok AI Video Generation for Every Creator
Create vertical or square concepts for TikTok, Instagram Reels, and YouTube Shorts, including motion and sound directions in the same prompt.
Prototype promotional clips, product scenes, and ad concepts before committing to a full production workflow. Review brand, product, and legal details before launch.
Use Grok for pre-visualization, mood clips, moving storyboards, and camera experiments before producing or animating a final sequence.
Turn product photos into moving visual concepts for listings, presentations, or campaigns, then verify product shape, text, color, and claims.
Everything About the Grok AI Video Generator
The Grok video generator creates short AI videos from text prompts or reference images. This page supports text-to-video and image-to-video, prompt-directed motion and camera behavior, optional native audio, and several duration and format choices.
The standard Grok video model on this page supports 6–30 second clips at 480p or 720p. Text-to-video supports 16:9, 9:16, 1:1, 3:2, and 2:3; image-to-video uses the uploaded images to guide the frame.
Grok can generate audio with the video based on sound directions in your prompt, including ambience, effects, music, or dialogue. Results vary, so preview the full clip and revise or edit important timing and speech when needed.
Use text-to-video when you want the prompt to define the whole scene. Use image-to-video when existing faces, products, characters, colors, or compositions should guide the result. Both workflows let you direct motion and sound in the prompt.
Commercial use depends on the applicable model, platform, and account terms, plus the rights attached to your prompts and reference images. Check the current terms and clear any people, brands, music, or source assets before publishing.
The standard Grok video form exposes Normal, Fun, and Spicy modes. They provide different creative directions; availability and output are still subject to the provider's safety rules and the content filter.
Describe the subject, action, camera movement, environment, lighting, mood, and sound. Use clear references when visual continuity matters, begin with a shorter clip, and change one prompt variable at a time when refining.
Text-to-video uses a written prompt. Image-to-video accepts up to seven reference images plus a prompt that describes motion, camera behavior, and sound. References guide the result but do not guarantee exact identity or product fidelity.
Yes. You can try the Grok Video AI Generator for free after signing in. Paid options are available when you need more generations, longer clips, or higher-resolution output.
Yes. Choose image-to-video, upload one or more reference images, and prompt the action, camera movement, mood, and sound. The images guide the subject and look, but important details should be checked throughout the generated clip.