FLUX 3 Max Product Video Prompts: Hero, Motion, Sound

The best FLUX 3 Max product video prompts start from your real packshot in image to video, then describe one motion and the sound it makes: a cap twisting, a pump clicking, a sole hitting tarmac. For hero shots without a packshot, text to video suits unbranded forms, with your logo and text added after export.

Its developers describe FLUX 3 Max as tuned for stronger prompt adherence and better visual quality. For product ads, adherence is the claim to check first: does the clip show the motion you wrote, and does the sound land on it? Every clip on VIDEO AI ME comes back with audio generated in the same pass, so the prompt has to direct both. The base structure (subject and action, camera, light and look, sound) is in the FLUX 3 Max prompt guide. This post applies it to products.

Why start FLUX 3 Max product video prompts from a packshot?

Text to video builds a product out of words. It can render an amber bottle from a description, but it cannot know your bottle's exact shape, label or cap. Image to video starts from your photo instead: the image becomes the first frame, the prompt drives the motion, and the audio is still generated. Because the photo carries the look, the prompt describes only what moves and what is heard.

Small on-screen text and exact logos inside generated frames are unreliable on every video model, so watch the label through the whole clip, not just the first frame, before it runs. Add prices, offers and logo lockups after export, in your own editing tool.

A packshot that gives the model something to work with:

• Sharp and evenly lit, with the product filling a good part of the frame.
• Room around the product, because a push-in, an orbit or a hand entering the frame needs space to happen.
• A surface that matches the ad: marble, linen, wet stone, a kitchen counter.
• Already the shape you deliver. With the ratio on auto, the clip follows your image; a named ratio overrides that.
• No text, badges or stickers added to the photo. Put those on after export, in your own editing tool.

The full mode walkthrough is in how to use FLUX 3 Max image to video.

Which motions work for which products?

Pick one motion per clip and write the sound that belongs to it.

The camera move should serve the motion: a slow push-in toward a pour, a half-orbit around a bottle, locked-off for a drop. The phrasing for each move is in the FLUX 3 Max camera movement guide.

How do you write sound on the action?

Write the effect and the action in the same phrase: "a sharp hiss as the tab cracks open", "the pump clicks as her thumb presses down". Tying each effect to a visible action and naming its source is what makes a product shot sound physical. It matters on paid social too: Meta's Reels ads page notes that Reels play with sound on by default and recommends vertical video with good sound.

Then decide the rest in order:

1. Voice. A product shot often needs none: write "no voice". If you want a line, mark it: A calm female voiceover says, "...", and keep it short. At roughly two to three words per second, an 8 second clip holds about 16 to 20 words, and a product beat usually wants fewer.
2. Ambience. It tells the viewer where the product lives: "kitchen room tone", "quiet studio room tone", "rain on the window".
3. Music. Name the style and where it sits: "a low electronic pulse", "soft acoustic guitar low in the mix", or "no music".

Audio is always generated. If your brand has its own track, still prompt the effects, then replace the music with your own audio afterwards.

Which FLUX 3 Max product video prompts work from a packshot?

These are image to video prompts. Attach the packshot, set the ratio, paste the prompt, and change the nouns to match your product.

1. Drink can, tab crack (packshot of the can on wet stone)

A hand enters from the right, lifts the can off the wet stone and cracks
the tab, and a fine mist escapes from the opening. The camera pushes in
slowly. A sharp crack and hiss as the tab opens, a soft fizz, a low
electronic pulse, no voice.

auto, 8 seconds, 1080p: 232 credits.

2. Serum dropper (packshot of the bottle with its dropper)

A hand lifts the dropper from the bottle and a single golden drop falls back
into the neck. The camera holds close and drifts slowly to the left. A faint
glass tap as the dropper lifts, a soft drip, quiet studio room tone, a gentle
ambient pad low in the mix.

9:16, 8 seconds, 720p: 136 credits.

3. Sneaker step (packshot of the shoe on concrete)

A foot slides into the shoe and presses down, and the sole flexes against
the concrete as a little dust lifts around it. Locked-off, low angle. A
heavy rubber thud, grit scattering on concrete, a single deep bass hit as
the weight lands.

auto, 5 seconds, 1080p: 145 credits.

4. Earbuds case (packshot of the open case on a desk)

A hand lifts one earbud out of the open case, then the lid snaps shut. Slow
push-in, close-up. A soft click as the earbud lifts out, a crisp magnetic
snap as the lid closes, quiet room tone, no music.

1:1, 8 seconds, 1080p: 232 credits.

What are good text to video product hero prompts?

When you have no packshot yet, or the product is a generic form, text to video can carry the whole shot. Describe the product's material and colour precisely, keep it unbranded, and add your logo after export.

5. Face cream on wet slate

A frosted glass jar of face cream sits on wet grey slate as a fingertip
scoops a small amount from the surface and lifts away. Macro close-up, slow
push-in. Soft diffused light from above, cool clean grade, gentle highlights
on the glass. A soft squish as the fingertip lifts the cream, water dripping
off the slate, a calm ambient pad, no voice.

4:3, 10 seconds, 1080p: 290 credits.

6. Pour-over coffee

Hot water spirals from a gooseneck kettle onto fresh coffee grounds in a
glass pour-over, and the grounds bloom and bubble. Close-up, locked-off.
Early morning light from a window on the left, warm grade, steam catching
the light. A warm female voiceover says, "Two minutes of patience. Worth
every one of them." The trickle of water, kitchen room tone, soft acoustic
guitar.

9:16, 12 seconds, 720p: 204 credits.

7. Blender in use

A hand drops frozen berries into a compact blender jar on a kitchen counter,
presses the base, and the jar fills with a swirling purple blend. Medium
close-up, locked-off. Bright morning light from a window behind, fresh
saturated grade. The thud of berries hitting the jar, a rising motor whir,
kitchen room tone, upbeat electronic music low in the mix.

9:16, 8 seconds, 720p: 136 credits.

8. Wristwatch at night, 21:9

A man in a navy overcoat checks a steel wristwatch as he steps out of a
black car onto a wet city street at night. Low angle, slow tracking shot
alongside him. Neon reflections in the wet street, deep blacks, teal and
amber grade. Rain on the car roof, a car door shutting, distant traffic on
the left, a slow pulsing score.

21:9, 10 seconds, 1080p: 290 credits.

Which settings suit product ads?

FLUX 3 Max costs 17 credits per second at 720p and 29 credits per second at 1080p, the same in both modes.

Test at 720p, deliver at 1080p. A 1080p run is a fresh generation, so keep the prompt identical to the draft you approved.

How do you turn one product into a test batch?

Shopify's product video best practices list a strong opening hook, a solution to a customer pain point, useful product details and a call to action as core elements. In a FLUX 3 Max clip, the hook is the motion in the first second, the detail lives in the sound and the voiceover line, and the call to action goes on as text after export.

A simple batch for one product: one packshot, three motions, two sound treatments (voiceover and no voice). That is six 8 second clips at 720p, or 816 credits. Take the best two to 1080p for another 464 credits, and the batch totals 1,280 credits, inside the Starter plan's 1,400 monthly credits. Plan allowances are on the pricing page, and the wider workflow for catalogues is in FLUX 3 Max ecommerce product videos.

Frequently Asked Questions

How do you write a prompt for a product video?

Describe one motion the product makes or undergoes, one camera move, the light, and the sound tied to the action, such as a hiss as the tab cracks open. With a packshot attached in image to video, skip the appearance and write only the motion and the sound.

Will FLUX 3 Max keep my product looking like my photo?

In image to video your photo is the first frame, so the clip starts from your real product. Check the label and shape through the whole clip before it runs, because exact logos and small text inside generated frames are unreliable on every video model. Add logos and text after export, in your own editing tool.

Should I put my logo or price in the prompt?

No. Keep the prompt about picture and sound. Add the logo, price and offer after export, in your own editing tool, where the text stays sharp and you can change it without paying for a new generation.

How long should a product video ad be?

One product beat usually fits in 5 to 10 seconds. FLUX 3 Max generates 5, 8, 10, 12, 15 or 20 second clips in one generation, and longer ads are built by joining clips in the editor.

Does FLUX 3 Max add sound to product videos?

Yes. Audio is always on, generated in the same pass as the picture, including effects, ambience, music and any voiceover you write. To use your own track, replace the audio with your own audio afterwards.

Can I use several product photos in one FLUX 3 Max video?

No. FLUX 3 Max on VIDEO AI ME takes one reference image, used as the start frame. If a brief needs several reference images in one clip, Seedance 2.5 and MiniMax H3 Max accept multiple references.