AI Models
What each model is good at, where it differs from the others, and what it costs to run.
xAI's versioned Grok video model. A short clip from a Visxy prompt or one still — not the unversioned Video row.
Credits come with a Generator plan. The exact cost is shown on the Generate button before anything is spent.
Grok Imagine 1.5 is xAI's versioned video model — not Grok Imagine Video, and not a stills row. On Visxy it is a video-library sting: 9:16, one idea, short enough to throw away.
xAI and third-party hosts talk about lip-sync, Fun/Spicy, and extending a clip. None of that is on this composer. Prompt, frame, size, length. No audio chip.
A short clip from a sentence of motion. Every second in this row's legal range is a real length, including a single-second punch-in this family's unversioned video SKU will not do.
9:16 is on the list. Image-to-video takes the frame from the still (no picker). Crop first.
What it is not: the unversioned Video row, Imagine 1.0, or grok.com's spicy playground.
Against Grok Imagine Video: own family. Video is the longer unversioned take, and it will not go as short as a sting. 1.5 is the preview clip. If the prompt needs more time than 1.5 can give, re-run the same motion on Video — do not pad this prompt with empty seconds.
This is not Grok Imagine or Grok Imagine 2.0. Those make stills. Animate a still by attaching it here or on Video.
One action. Paste the library prompt; delete the second location.
Short first. Add your own track if the clip needs sound.
No Visxy run yet.
No public Grok Imagine 1.5 Edit page. Attach one still. Describe only the motion.
No. 1.5 is the versioned short clip. Grok Imagine Video is the unversioned, longer SKU.
Not on this row. That length belongs on Grok Imagine Video, inside that model's range.
No audio control on this row.
Attach it to Grok Imagine 1.5. No separate Edit article.
