Imagine Video 1.5 with References release notes

xAI has introduced updates to Imagine Video 1.5, adding text-to-video capabilities, native 1080p resolution, and the ability to use image and voice references to maintain consistency across generated videos.

Text-to-Video and Native 1080p Support

Imagine Video 1.5 now supports native 1080p generation for both text-to-video and image-to-video workflows. The text-to-video feature allows users to generate video content from a text prompt alone, removing the requirement for a starting image. This is achieved by pairing xAI's image generation with image-to-video technology.

These features are generally available on grok.com/imagine, iOS, and Android platforms.

Voice and Character Consistency

Imagine Video 1.5 enables the use of image and voice references to ensure consistency across different scenes. By providing a character image and a voice reference, the model maintains the same facial features and voice for a character throughout the video generation process.

Multi-Reference Control

The model now supports up to seven references per generation. This allows users to lock specific elements—such as a face, a product, or a location—while varying other parameters. Users can:

• Keep a character and change the scene.
• Keep the scene and change the character.
• • Hold both the character and scene and change only the action.

API Availability and Integration

Image references, text-to-video, and native 1080p are currently live in the xAI API via the grok-imagine-video-1.5 model. Voice reference support is available on request through xAI sales.

API Implementation Example

Developers can implement video generation using the following Python SDK structure:

import os
import xai_sdk

client = xai_sdk.Client(api_key=os.getenv("XAI_API_KEY"))

response = client.video.generate(
    prompt="Slow cinematic push-in as embers drift across the battlefield and the helmet's crest stirs in the wind",
    model="grok-imagine-video-1.5",
    reference_image_urls=["https://example.com/helmet.jpg"],
    duration=6,
    aspect_ratio="16:9",
    resolution="720p",
)

print(response.url)

Availability

Image and voice references are available starting July 31, 2026, in the US for SuperGrok Heavy and SuperGrok Plus users on grok.com/imagine and iOS, with a rollout to all tiers following in the subsequent days.