AI Product Photography Prompts: A Working Method

Updated

An AI product photography prompt needs four things: the product described by material and finish, the surface and background it sits on, the lighting setup, and the camera angle. Add empty space if the image will carry text, and upload a photo of the real product when accuracy matters.

The template

[PRODUCT] made of [MATERIAL / FINISH], placed on [SURFACE] against [BACKGROUND]. Lighting: [SETUP, e.g. large softbox from the left, subtle rim light from behind]. Camera: [ANGLE], [LENS], shallow depth of field. [PROPS, if any]. Leave empty space on the [SIDE] for text. Commercial product photography, [FORMAT].

Describe materials, not quality

"High quality" tells the model nothing. "Frosted glass bottle with a brushed aluminium cap" tells it how light should behave. Name the material and its finish for the product and for the surface.

Lighting setups that work

LookHow to describe it
Clean e-commerceEven soft light, white seamless background, soft shadow under the product
Premium / luxuryDark background, single side light, strong rim light, glossy reflections
Fresh / naturalDaylight from a window, linen or wood surface, soft shadows, plants out of focus
DynamicSplash or levitating elements frozen in mid-air, hard light, high contrast

Angles

  • Eye level: bottles, boxes, anything with a front label.
  • 45 degrees: the default for most products, shows top and front.
  • Top-down: flat lays, cosmetics sets, food.
  • Low angle: makes a product look larger and more imposing.

Keeping the real product accurate

A text prompt invents a product. To show yours, upload its photo and ask the model to keep the product, label and proportions unchanged while replacing the scene around it. Check the label afterwards: small text is where errors appear.

Prompts to study

Real prompts from the network, each published with the result it produced:

More guides