P—07 / ESPRESSO, MACRO

Product macro prompt for MiniMax H3 Max

A product shot where the product is a liquid, and the sound does half the work.

Four hundred characters of commercial macro work. It picks a subject that changes shape while you watch, lights it through a window rather than with a rig, keeps the camera slightly alive, and then spends a quarter of its length on sound — which for a shot about pressure and heat is where most of the appetite lives.

Steps
One shot
Engine
MiniMax H3 Max
Mode
Text to Video
Duration
5s
Ratio
16:9
Resolution
768P
Audio
Machine foley + room tone
Cost
$0.40

Reference render — not generated on this site. Source: fal.ai

Output reference

Espresso runs from a chrome portafilter into a glass, golden crema blooming, a cafe out of focus behindVideo
The clip
389 / 7000
  • Free clipThis run spends credits — the free clip is 5s · 480P · 16:9.

1 FREE CLIP · NO AUDIO

01 — Inside

One prompt, one workflow

The clip beside the verbatim source. Copy it into the console above, change what you need, generate.

Reference output

Espresso runs from a chrome portafilter into a glass, golden crema blooming, a cafe out of focus behindVideo
The clip
The promptthe clipMiniMax H3 Max · Text to video
389 chars

The pull

Subject and transformation, the light, three imperfections, the lens and its drift, then four sounds.

Macro shot of espresso pulling from a chrome portafilter into a warm glass, thick golden crema blooming and swirling. Bright morning light through a cafe window, steam curling, water spots on the chrome, a few coffee grounds on the tray. 100mm macro, shallow focus, slight handheld drift. Sound: grinder winding down, the hiss of the pump, liquid pattering into glass, cafe chatter behind.

4 levers

Make it yours

What is safe to change. Most libraries publish only this list, which is why so many copied prompts come back worse than the original.

  1. 01

    The transformation

    Crema blooming and swirling is the shot. Any product that visibly changes during the clip works the same way — chocolate setting, ink dispersing, butter melting. A static product needs a moving camera instead, and that is a different prompt.

  2. 02

    The light source

    Bright morning light through a cafe window is a direction, a colour temperature and a reason. Swap it for a single hard source or an overcast wash and the chrome behaves completely differently, because chrome shows you the light rather than the object.

  3. 03

    The imperfections

    Steam curling, water spots on the chrome, a few grounds on the tray. Three small pieces of evidence that a person works here. Replace them freely; removing all three is what makes product renders look like renders.

  4. 04

    The handheld drift

    "Slight handheld drift" keeps a macro shot from looking locked to a tripod in a studio. It is a small amplitude on purpose — macro plus real handheld movement is unwatchable.

Three ways to break it

  • Asking for a logo or readable branding

    This prompt names no brand, and that is why it works as a template. Rendered lettering is legible when you give an exact string to set; a brand the model has to invent comes back as approximate type on a real-looking cup.

  • Dropping the shallow focus

    "100mm macro, shallow focus" is what makes it macro. At a deep stop the same scene is a photograph of a coffee machine, and every one of the small details becomes clutter instead of texture.

  • Replacing the four sounds with a mood

    Grinder winding down, pump hiss, liquid pattering, cafe chatter behind. Four sources at three distances. "Cafe ambience" is one source at one distance, and it costs the shot its depth.

What it does

What the espresso, macro template does

Product work is where generated video is most often obviously generated, and this prompt avoids nearly all of the usual tells. Reading how is more useful than reading what it depicts.

It starts by choosing a subject that does something. Espresso pulling into a glass with crema blooming and swirling is a transformation with a beginning and an end, which means five seconds has an arc rather than a duration. Most product prompts describe an object and then have to invent a camera move to give the clip somewhere to go; this one does not need one, and the camera stays almost still as a result.

The lighting choice is the second good decision. Bright morning light through a cafe window is a motivated source — there is a reason for the light and it is in the world of the shot. Chrome is a mirror, so a portafilter rendered under unmotivated studio light shows you reflections of nothing, and reflections of nothing is one of the most reliable ways a viewer knows something is synthetic without being able to say why.

Then three imperfections in a row: steam curling, water spots on the chrome, a few coffee grounds on the tray. None of them is the subject. All of them are evidence that the machine is used. The water spots in particular are doing the work of a hundred words of description, because a spotless chrome surface is the default a render produces and a spotted one is not.

The camera is described in six words — 100mm macro, shallow focus, slight handheld drift — and each does one job: how close, how much is sharp, how alive. That is a complete camera specification for a shot like this, and adding to it would take away from the subject.

A quarter of the prompt is sound, and the mix is layered by distance. The grinder is winding down, so it is already finishing as the clip starts, which places the shot in time. The pump hiss and the liquid pattering are at the subject. The chatter is behind. Four sources at three depths, generated in the same pass as the picture — on a silent model this shot needs a separate sound design brief and a sync pass to reach the same place.

Everything here runs in text to video, on MiniMax H3 Max.

3 questions

Espresso, macro — common questions

  • 01

    Can I use this for my own product?

    Swap the subject for something that changes during the clip and keep everything else. If your product does not change, add a slow move with a named amplitude and speed instead.

  • 02

    Will branding come out legible?

    On-screen text is more reliable when you give the exact string to set. A brand the model has to invent will be approximate. For real packaging, reference input is the tool, and that runs on MiniMax H3.

  • 03

    Why 16:9 for a product shot?

    Only because that is the showcase framing. Text to video takes 21:9, 16:9, 4:3, 1:1, 3:4 and 9:16, and the ratio has to be a concrete one — adaptive is refused.

Copy it, change three things, run it.

Every character is on this page. 5s at 768P costs $0.40.

Generate free · queue

Written and maintained by the MiniMax H3 Max AI Video Generator editorial teamPublished Last updated