xAI launches Grok Imagine Video 1.5 with better audio sync and faster generation

xAI, led by Elon Musk, has officially launched the Grok Imagine Video 1.5 model. It can turn a static image into a short video with synchronized audio. Compared to the previous version, it brings better audio sync, more realistic motion, and faster generation. The model outputs sound effects, ambient noise, and dialogue all at once, making lip-sync more natural for character dubbing. It also reduces common issues like distorted limbs or floating objects, and better simulates weight and momentum. In Fast mode, generating a 6-second 720p video takes about 25 seconds, down from over 40 seconds before. The model is now available in the xAI API as grok-imagine-video-1.5. Developers can upload a starting image, describe the motion, and choose resolution and duration to use it.