MB SuiteMB Suite

Photo Session: product photos with AI

Updated 2026-08-27

In two lines: in Social Media > Content Studio > Photo Session you upload a photo of your product and the AI shoots it again: same product, professional set, your style, in the formats you publish in. It's the closest thing to a photo shoot without booking a studio.

Use it when what you need is the product, not a post: catalog photos, launches, a marketplace listing, images to feed later into an ad or a series. If what you want is a batch of posts with copy and a calendar, that's Content Studio: generate content with AI; if it's ads for a campaign, it's Create creatives and copy for your ads with AI (Ad Creator).

How do I set up a session?

In the Photo Session tab, click New Session, name it, and go through the three steps:

  • Product — choose the Mode: Single product (one product, several photos of it) or Multi-product (up to 20 products sharing the same style, ideal for a catalog). Then upload the product photos, or pick them from your Library.
  • Creative — pick the Session type (Studio, In scene, In action, Flat lay, With model or Detail); depending on the one you choose, a box appears to describe the scene, the action, the model or the detail to highlight. Then set the Reference: Free (no reference, the AI decides), Inspire (uses the mood of your reference photo) or Replicate (copies its composition, angle and lighting exactly — only the product changes). Last, the Logo: Logo, Icon or No logo, with its position.
  • Output — choose one or more Formats (Feed (1:1), Portrait (4:5), Story (9:16), Landscape (16:9), Ultra-wide (21:9)) and, in single-product mode, the Number of photos. Each format multiplies the total: the wizard shows you the resulting count before you launch.
  • Create & generate — the session is created and generation starts on its own.
For the first session, go small: one product, two photos, one format. Look at the result, adjust the type or the reference, and only then launch the whole catalog. Calibrating with two photos costs a fraction of what it costs to redo twenty.
The photo you upload is what the AI has to recognize the product from: a clear picture, well lit and with the product fully visible, changes the result more than any description. With Replicate the reference photo also matters more than usual — the AI copies its framing, so if the reference is a close-up, all your photos will be close-ups.

The generation runs on its own

Each image can take 2 to 5 minutes, so a session with several products and formats takes a while. You don't have to stay and watch: the session shows Generating, you can leave the view and keep working, and we send you an in-app notification when it's ready — the notification takes you straight back to it.

If an image fails, the session ends as Partial: the credits for the failed ones are returned and the Retry button relaunches just what's missing.

Create variant: change a detail… or the format

You don't have to redo the whole session because of one photo. Hovering over any generated photo shows the Create variant button, which opens a dialog with two controls you can use together or on their own:

  • What do you want to change? — describe only the change ("make the ring a transparent diamond instead of yellow gold"). Everything else stays as it was.
  • Variant format — the base photo's format comes preselected. Pick another one and the AI reframes the same scene into it: same product, same lighting, same style, recomposed natively for the new proportion.
  • Generate variant — the new photo appears right next to the one it came from, saved with its own format.

When you change the format, the instruction becomes optional: "the same one, but for story" is already a complete request. And it's a real reframe, not a crop — the AI recomposes the shot for the new proportion instead of stretching the image or padding it with bars, so the story doesn't come out with the product cut in half.

This is what lets one shoot cover every placement: generate the session in the format you use most, and from the photos you actually liked ask for the variants for story, portrait or landscape. You only pay for the framings you're going to use, not for every combination.

What does it cost in credits?

Every generated photo consumes Multimedia credits — the session's own photos and the variants alike. If an image fails, its credits come back. To see how many you have left and how the budget is split by category, see See your AI usage and control your credits.

Where do the photos end up?

Every photo of the session —variants included— is registered in your Media Library, in a folder named after the session, so you can use it in a post, in an ad or in a report without downloading anything. From the session itself you can also use Download all, or download and delete photos one by one. See Organize your files in the Media Library.

Deleting a session deletes all the photos generated in it, and there's no undo. If you're only after space, delete the photos you're not going to use one by one and keep the session.
If your session involves people —the With model type, or a reference photo where someone appears— you are responsible for holding the rights and the consent to use those images. Uploading and processing that material is governed by the Terms & Conditions.

Still have questions?

Ask MIA, the MB Suite AI assistant: open it with the MB AI button (⌘J) in the top bar. MIA knows the section you're in, so you can ask it about what you see on screen — and it can also answer how-to questions about using MB Suite.