June 18th xA The company issued an announcement yesterday, June 17thMake it official Grok Imagine Video 1.5 Model, and closes the preview formally online in xAI API。
xAI launched the Grok Imagine Video 1.5 Preview preview on 3 June 2026, with the main hit "Step Chart One Generates Short Videos with Synchronized Audio" and, compared to the previous generation, the main upgrades to the synchronization, motor authenticity and production speed。
With regard to the synchronization of acoustics, the model produces both sound, ambient and conversational in a single generation, aligning accurately with audio and image actions. The model greatly enhances voice clarity and is more natural in oral synchronization, and applies to short film creations that require the voice of people。
In terms of physical and physical expression, Grok Imagine Video 1.5 enhances motor coherence in the video, significantly reducing common physical distortion and object floats. Models can better simulate the sense of weight and momentum, such as the natural shaking of a person ' s walking clothes, the acceleration trajectory of a falling object, and a sense of wholeness closer to the real physical world。
In terms of the speed of generation, in the Grok Imagine Video 1.5 Fast mode, it takes about 25 seconds to generate a video with a resolution of 720p, compared to 40 seconds or more for the previous generation。
Grok Imagine Video 1.5 has completed the preview and is provided in xAI API with the gok-imagine-video-1.5 model name. Developer can call by uploading the initial picture, describing the mode of movement, and choosing resolution and duration。