AI Models
What each model is good at, where it differs from the others, and what it costs to run.
xAI's unversioned Grok video model. A longer clip than 1.5 — not the stills rows, and not a five-second sting.
Credits come with a Generator plan. The exact cost is shown on the Generate button before anything is spent.
Grok Imagine Video is xAI's unversioned video SKU — not Grok Imagine 1.5, and not the stills pages. On Visxy it is a video-library take: 9:16, one action, a clip that is allowed to run longer than 1.5 can.
It will not go as short as 1.5. A one-second sting belongs on 1.5. This row's floor is a real beat, not a blink.
Fun/Spicy, Extend, and reference-to-video are not on this composer. Mode is normal. Prompt, frame, size, length. No audio chip.
A followable action that needs more than a preview clip. 9:16 from the library. Image-to-video takes one still and inherits its frame — crop first.
What it is not: 1.5 with the version number filed off, Grok Imagine with motion, or a talking-head studio.
Against Grok Imagine 1.5: own family, own ceiling. 1.5 is the short Grok clip (and the one that can start at a single second). This page is the longer unversioned take. Animating a 1.5 draft must not become a Video job — on Visxy it will not: they are paired separately.
Against the stills rows: Grok Imagine and Grok Imagine 2.0 make pictures. Animate those pictures here (or on 1.5), do not generate the still on a video model.
One shot. Paste the library prompt; delete "then."
9:16 from the library. Use the extra length on purpose — empty seconds drift.
No Visxy clip yet. Generate fewer.
No public Grok Imagine Video Edit page. Attach one picture. Describe only the motion. No last frame.
No. Separate catalog row. 1.5 is the short, versioned clip. This is the unversioned video SKU.
Not on this row. A sting that short belongs on Grok Imagine 1.5.
No. Normal mode only. No audio control. No Extend.
Attach one picture to Grok Imagine Video. No separate Edit article.
