
A new feature from GROK
😕 Standard Grok Imagine = a generator.
You write a prompt → it creates an image/video → then you edit it yourself.
⭐️ Grok Imagine Agent = workspace + agent.
You give it a task → it plans, generates, edits and moves objects on the canvas itself, and can iteratively assemble a scene or video.
I gave it a task as an example, asking it to come up with different scooter designs in different locations; of course, I meant to customise a Honda ADV 160, but it suggested its own options. Take a look at the quality. What I liked most was how it made a video where my updated ADV is riding through the streets of Tokyo



English













