---
name: animate-props
description: "Keep an unusual prop (glasses with three lenses, a strange tool) right in an AI video — words fail (the model normalises the object, \"always\" glues it on, describing where it sits adds a floating extra, forbidding the mistake brings it back); the stable recipe is GPT Image 2.5 still of the character wearing/holding the prop → GPT start frame in the location → MiniMax H3 Max Turbo image-to-video with BEGINNING / THEN / END, 7 s 768P. Use when the user says the prop changes, wrong object, keep the prop, my character holds / wears. Part of the animate meta-skill."
---

These creative references describe source experiments with external tools. Use your own supported account and tools; this download installs no automation. Settings and timings are experimental observations, not current FlashBoards availability.

# animate-props · a prop that holds its shape

> 🏆 **Source quality recommendation:** Seedance 2.5 reference-to-video (`bytedance/seedance-2.5/reference-to-video`, references as @Image1… / @Audio1…, 4–30 s, native sound) is the supplier's preferred quality route. That recommendation is supported by the product-commercial comparison, not a measurement of every character task. The H3 methods below retain their own measured inputs, settings and timings. Use supported external tools; this is not a FlashBoards integration claim.

A video model has seen a million ordinary objects and pulls an unusual {prop} back to the usual one. Tested prop: one frame with **three** round lenses in a row (a product shot, 400 × 400).

## Choose

| Priority | Recipe | Time / tested output | Result |
|---|---|---|---|
| 🏆 the one that holds | GPT still of {character} with the {prop} → GPT start frame → **H3 Max Turbo image-to-video** · 7 s · 768P | queue 2 s, run 33.4 s | three lenses from the first frame until the hands take them off; then in the hand to the last frame |
| ⚡ ordinary prop | Image 3 in `animate-multi-image` | 5 s 480P | fine for an ordinary object only |

## Steps

1. **The prop on the character** — GPT Image 2.5 Sunburst edit, quality medium, 1536×1024; picture 1 = the person's {character sheet}, picture 2 = the product shot of the {prop}. The prop is described in words here (an image model, not the video model); the person is not:
   ```
   Picture 1 is {character}. Picture 2 is {a prop}. Photograph {character} from picture 1 {wearing / holding} exactly the {prop} from picture 2 — the same {unusual feature, its count in CAPITALS, e.g. THREE round lenses in one row}, the same {material}, the same {colour}, the same {other details}. Front view, head and shoulders, looking into the lens, soft even light, plain dark grey background. Face, hair and clothes exactly as in picture 1.
   ```
2. **The start frame** — GPT Image 2.5, quality medium, 1920×1088; picture 1 = the {location}, picture 2 = the still from step 1; frame it so the prop is big enough to read:
   ```
   Picture 1 is a location: {location in a few words}. Picture 2 is {character} {wearing / holding} {the prop in a few words}. Photograph {character} from picture 2 {standing} in the {location} of picture 1, a medium shot from the waist up, facing the camera in the middle of the frame, the {location} stretching behind, the same {surfaces, details} and {light} as in picture 1. {character} {wears / holds} exactly the {prop} of picture 2 — the same {unusual feature}, the same {material and colour} — and exactly the face, hair and clothes from picture 2. Cinematic film still, 16:9.
   ```
3. **The video** — `minimax/h3-max-turbo/image-to-video`, `{image_url: <start frame>, duration: 7, resolution: "768P", prompt_expansion_mode: "disabled"}`. The action in three steps, nothing says "always", the prop is never counted or forbidden:
   ```
   Live-action film shot, one continuous shot: {character} stands in {location}, medium shot from the waist up, facing the camera. The video begins exactly on the given first frame and is alive from it. The camera slowly pushes in. {background motion}; {light}.
   The {prop} stays exactly as in the first frame while {character} {wears / holds} it: the same {unusual feature}, the same {material and colour}.
   BEGINNING: {character} stands still {wearing / holding} the {prop}, looking into the lens.
   THEN: {the action with the prop}.
   END: {where the prop is until the last frame}; {what is bare / free}, and {character} looks straight into the lens.
   Exactly one person is in the frame: {character}. Sound: {sound}.
   ```

## Never (all H3 Max reference-to-video, same three images and seed 20260929, 480P)

| Tried | Result | Duration |
|---|---|---|
| Only the product shot: "Image 3 is her glasses." | **two lenses** — the object normalised to ordinary glasses | 5 s |
| Describe it with "always": "Count them: three lenses, always." | three lenses — **and they never leave**: the character takes them off and they are back on the face | 5 s |
| Describe where it sits: "the third lens … sticking out on the RIGHT side of the frame" | **a fourth lens floats in the air** beside the head | 7 s |
| Forbid the mistake: "no fourth lens, no loose lens, nothing floating in the air" | **back to two lenses** — a negation names what you do not want and pushes the count down | 7 s |
| The prop-on-character still from step 1 as Image 3 | in a wide 480P shot the glasses are a few pixels across and read as two lenses — a reference helps less than a first frame | 7 s |

## Example run (reference only — never reuse its files, prompts, faces, voices or frames for a person)

The demo character Mara Vale with glasses of three round lenses in one row (thin gunmetal wire, pale mint-green glass, temples with black tips) in a flooded tiled corridor (teal tiles, water on the floor, warm side light). Video action: BEGINNING she stands still wearing the glasses; THEN lifts both hands, takes them off by the temples and lowers them to her chest; END they stay in her right hand at her chest, face bare, eyes into the lens. Sound: dripping water and a low hum.