Skip to main content

Grok Imagine 1.5 reference to video with a character pack

Grok Imagine 1.5 (grok-imagine-video-1.5) has a reference-to-video mode: you give it images of a character and it keeps that character in a new scene without locking the first frame. A turnaround sheet already has the front, side and back views it needs. Here is which pack file to upload in which order, and how to point at them in the prompt.

We have not made an example with Grok Imagine 1.5 yet. This page maps the pack to xAI's published limits; the examples on this site were made with Kling 3.0.

What Grok Imagine 1.5 allows

Checked 2026-10-22
  • Up to 14 reference images per request on grok-imagine-video-1.5 (up to 7 on grok-imagine-video).
  • Reference-to-video clips run up to 15 seconds at up to 720p.
  • In the prompt, refer to images by position, starting at <IMAGE_0>. If you also pin a first frame with image, it takes <IMAGE_0> and the references start at <IMAGE_1>.
  • Optional: up to 3 preset voice references with lip-synced speech, a pinned first or last frame, and up to 4 keyframes inside the clip.
  • xAI's advice for many references: give each one a single role and a fixed place in the scene, keep the cast small and keep props static.

Sources: xAI Docs: Reference-to-Video · xAI Docs: Video Overview

Upload order

Upload in this order so the tokens in your prompt stay the same for every shot. The first slot is <IMAGE_0>.

  1. <IMAGE_0>02_turnaround_views/view_front.pngFace and front of the outfit
  2. <IMAGE_1>02_turnaround_views/view_side.pngProfile and hair from the side
  3. <IMAGE_2>02_turnaround_views/view_back.pngHair and outfit from behind
  4. <IMAGE_3>01_character_reference/full_body.pngHeight and proportions
  5. <IMAGE_4>03_outfits/outfit_work.pngThe job outfit
  6. <IMAGE_5>04_props/prop_role.pngThe product or tool
  7. <IMAGE_6>05_scenes/scene_workspace.pngThe location

Seven images is plenty for one character. Add expressions or poses from 07_expressions_poses only when the shot needs them, and keep the total at 14 or fewer.

Why three separate views beat one good angle

A single image, even a well-chosen three-quarter view, only shows part of the character. When the character turns, the model has to invent the profile and the back, and that is where the face, hair and outfit start to drift. Upload the front, side and back as separate references and give each one a single role in the prompt, as the official guide recommends. If you also have a three-quarter view, add it as an extra reference rather than using it on its own.

Keep one face per image: crop expression grids out of the sheet, and paste the same locked appearance description into every shot's prompt.

Point at the references in the prompt

Prompt template

<IMAGE_0>, <IMAGE_1> and <IMAGE_2> are the front, side and back of the same character, [name]: [skin, eyes, hair]. Body proportions as in <IMAGE_3>. She wears the outfit from <IMAGE_4> and holds the [product] from <IMAGE_5>, in the room from <IMAGE_6>. [Action]. [Camera], single continuous shot. No extra people, no text or logos.

For a sequence, pin the last frame of one clip as the first frame of the next so the character carries over between cuts.

Read the full character consistency guide

Seedance 2.5MiniMax H3

Start with a free character

Every character's full 19-image pack is free: eight starter characters need no account, the rest a free sign-up. All nine example ads use pack images only.

Browse free characters