Example: several products in one setting
You have a range of three products and you want a group shot, in a brand setting, without organizing a photo shoot. With reference images, Seelab brings your products together in a single scene, with consistent lighting and shadows.

What you need
- A sharp photo of each product, ideally on a plain background. The training images of your Seelab models are perfect: they are available in the Resource Hub, Datasets tab.
- An image of the setting, empty. You can generate it first with Seelab ("travertine shelf in late-afternoon light, leaf shadows, no objects").
Step by step
- Choose the model and the format. Here, Nano Banana 2.1 in Banner 16:9.
- Add the products first. Image button left of the prompt, then Resource Hub › Datasets. Select one photo of each product: they become images 1, 2 and 3.
- Add the setting last. It becomes image 4. The counter shows 4 images out of 10.
- Write the prompt, giving each image a role and a position.
- Check the cost, then launch.

Place the three products from images 1, 2 and 3 side by side on the travertine shelf from image 4, well spaced: the mist from image 1 on the left, the serum from image 2 in the center, the oil from image 3 on the right. Keep their shapes and labels exactly. Keep the setting, the lighting and the leaf shadows from image 4, with realistic cast shadows under each bottle.
Why this prompt works
- Each image has a role: images 1 to 3 are products, image 4 is the setting.
- Each product's position is given: left, center, right, "well spaced".
- What must not change is stated: the shapes, the labels, the lighting of the setting.
If the result isn't right
- A product is distorted: run it again, or use a sharper, more front-facing photo of that product.
- The products overlap: specify the spacing and the relative size ("the serum slightly larger, in the foreground").
- You need perfect fidelity: build the scene with the Composer instead, which integrates each product without redrawing it. See Composer: create a composition.
Updated on: 31/10/2026
Thank you!
