grok's image2 is indeed powerful, and the scale is massive. Just one line: 'Generate a screenshot of an ancient ARPG MMO open-world game themed around *The Plum in the Golden Vase*.' I just compared it, and it's way better than the raw images from Gemini and grok. If you then input into grok: 'Generate a dynamic video based on this game screenshot, with the character walking forward in a seductive manner,' the result will be shocking. I don't even dare to post it on Twitter—I’m afraid my account might get banned. grok's video scale is also insanely bold...