Grok Imagine 1.5 AI Video Generator

AgentReel offers two video models from xAI: the original Grok Imagine Video, marked Fast, and Grok Imagine Video 1.5, which this page covers. Both take a text prompt or one start image, share the same five aspect ratios and the same three lengths, and cost the same per second at 480p and 720p. The original adds a 1080p setting; version 1.5 stops at 480p, 720p.

In practice, pick 1.5 when you want xAI's newer model for a draft or a phone-first clip, and switch to the original when the clip must be 1080p. Running one prompt on both, side by side, is the quickest way to see which result you prefer.

Describe a single shot: who or what is on screen, the action, the place, the camera and the light. "A street musician tuning a cello under a railway bridge at dusk, slow dolly in, warm sodium lights, light rain" gives the model a clear plan; "a cool city video" does not.

Use the length you picked. A 6-second clip holds one action, a 15-second clip can hold an action and a reaction or a slow reveal. Keep to one location per prompt and make sequences from several clips joined later in the AI video editor.

With a start image the clip opens on your picture, so do not repeat what is already visible. Write only the change: the steam rises, the dog turns its head, the camera drifts left. The clip follows the shape of the image, so crop it to the frame you want first, for example with the free image cropper.

A clean, sharp picture gives a cleaner clip. If the source is small or soft, run it through the free image upscaler before you upload it. To make the start image itself, use the AI image generator.

480p costs about half as much per second and is enough to judge motion, timing and framing. Make your drafts there. Render the keeper again at 720p for the version you post; for phone feeds that is usually all you need.

If the final has to play large, on a website header or a presentation screen, render it on a model with 1080p or 4K instead of stretching a 720p clip.

A clip costs 4 to 17 credits depending on resolution and length; the full table is on this page and the Generate button shows the exact amount before anything is charged. A failed task is refunded automatically.

Video models are paid from a plan or credit pack. The free credits of a new account cover Z-Image, video templates and Code to Video, not this model. The credits and pricing guide explains the difference between plans and packs.

The comparison table above lists every video model in the generator. For a larger frame, Gemini Omni 1.1 Flash and Veo 3.1 reach 4K; for sound with the picture and up to 1080p, look at Seedance 2.5 or Kling 3.0. The model overview shows every price range at a glance.