Image-first generation
The preview model requires an input image, which makes it best for animating a portrait, character, product image, scene still, or visual concept rather than generating from text alone.
Essential cookies keep the app working. Optional analytics, support, and marketing cookies help us improve the site and services. Cookie Policy.
Grok Imagine Video 1.5 Preview
This page is for users searching Grok Imagine Video 1.5 Preview directly. The model is designed for image-to-video generation: upload one source image, describe the motion, choose duration and resolution, then generate a short clip from the visual reference.
Credits
236
Input
Image + Prompt
ETA
1m
Duration
1s / 2s / 3s / 4s / 5s / 6s / 7s / 8s / 9s / 10s / 11s / 12s / 13s / 14s / 15s
Aspect Ratios
auto / 16:9 / 9:16 / 1:1 / 4:3 / 3:4 / 3:2 / 2:3
Audio
Silent
Built for image-to-video workflows where a source image is required and the prompt controls motion, camera direction, and scene behavior
Supports 1 to 15 second clips, 480p or 720p output, and common aspect ratios including vertical, square, landscape, and auto
A practical preview model for creators who want to test newer Grok Imagine video behavior from a single visual reference
Searches for Grok Imagine Video 1.5 Preview usually come from users evaluating a newer xAI image-to-video path. They need to know whether it requires an image, what controls are available, and how it differs from broader Grok Imagine workflows.
The preview model requires an input image, which makes it best for animating a portrait, character, product image, scene still, or visual concept rather than generating from text alone.
Choose a clip length from 1 to 15 seconds, then pair it with 480p or 720p output depending on whether you want lower cost or sharper detail.
The model supports auto, square, widescreen, vertical, 4:3, 3:4, 3:2, and 2:3 aspect ratios, so it can fit social clips, product previews, and character-focused videos.
Start with a clear JPG, PNG, or WebP image. This model requires image input and supports one image per generation.
Write a prompt that explains the movement, camera behavior, facial expression, atmosphere, or transition you want the image to become.
Use the duration slider from 1 to 15 seconds, choose 480p or 720p, then generate the clip from your source image.
Yes. This preview model is an image-to-video model. You need to upload one source image before generation.
The model supports durations from 1 to 15 seconds in one-second steps.
Grok Imagine supports broader text-to-video and image-to-video workflows. Grok Imagine Video 1.5 Preview is presented here as a newer image-to-video preview path with one required input image and more granular duration control.
These links help users move from model research into the exact workflow they want to try next.