Photoshoot API

AI product photography API

Send one product photo, even a flat lay on the floor, and say what you want. Garmo makes new photos: the product in a setting your prompt describes, or worn by a Garmo model, your own model or one Garmo picks, in planned poses.

Test keys are free and need no pack.

  • A prompt in any language, or none and Garmo plans the set.
  • On a model: Garmo’s models, your own, or "auto".
  • 1 to 4 images per request, each a different angle or pose.
Terracotta tiered sundress laid flat on a wooden floor
Input
A model wearing the terracotta sundress in a stone colonnade
Front
The same model walking in the sundress
Walking
POST https://api.garmo.studio/v1/generate
{
  "model": "auto",
  "prompt": "a sunlit stone colonnade, warm late-afternoon light",
  "count": 2
}

One request

Every request is asynchronous: the request returns a job in about two seconds, the images follow in 20 to 90 seconds by webhook or by polling.

curl https://api.garmo.studio/v1/generate \
  -H "Authorization: Bearer $GARMO_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -H "Content-Type: application/json" \
  -d '{
    "image_url": "https://shop.example/products/sundress-41.jpg",
    "model": "auto",
    "prompt": "a sunlit stone colonnade, warm late-afternoon light",
    "count": 2
  }'
Finished job · GET /jobs/{id}
{
  "id": "job_8Kq2xV7mPz4LtR9wNc3B",
  "object": "job",
  "tool": "generate",
  "status": "completed",
  "livemode": true,
  "outputs": [
    {
      "id": "out_1",
      "status": "succeeded",
      "label": "Front",
      "url": "https://….r2.cloudflarestorage.com/renders/2026/11/8f3c…0.webp?X-Amz-Signature=…",
      "url_expires_at": "2026-11-20T14:00:00Z"
    },
    {
      "id": "out_2",
      "status": "succeeded",
      "label": "Walking",
      "url": "https://….r2.cloudflarestorage.com/renders/2026/11/8f3c…1.webp?X-Amz-Signature=…",
      "url_expires_at": "2026-11-20T14:00:00Z"
    }
  ],
  "credits_charged": 2,
  "credits_refunded": 0,
  "metadata": {
    "sku": "DR-123"
  },
  "created_at": "2026-11-19T14:02:11Z",
  "completed_at": "2026-11-19T14:02:58Z"
}

Parameters

promptWhat you want, in your own words and any language (up to 500 characters): setting, light, props, mood. Without a model it decides everything, including whether hands or a person appear; with a model it is the art direction. Leave it out and Garmo plans the set.
countHow many images, 1 to 4. Each is a different angle or pose.
modelPut the product on a model: an id from GET /catalog/models (a Garmo model or custom_<id> for your own), or "auto" to let Garmo pick one. Leave it out for product photos without a model.
categoryWith a model: what the product is, so the poses fit.
sceneWith a model: the place. auto plans a setting that suits the product.
genderWith model "auto": who wears it.
aspect_ratioShape of the images.
Full reference

Cost

1 credit per image

1 API credit = 1 image. Failed and canceled images are refunded automatically.

API pricing

What people build with it

New products, ready to list

A product added to the shop gets on-model photos before anyone books a shoot.

Flat lays to on-model

Turn the photos suppliers send into a consistent set on the same model.

Seasonal refreshes

The same product in a new setting for every campaign, from one prompt.

Prefer to click? Use it in Garmo Studio →

Good to know

Questions, answered.

What developers ask before connecting Garmo to a shop or a workflow.

Still stuck? Email us

Do I need a prompt?

No. Without one, Garmo plans the set from the photo: a clear hero view first, then other angles and settings that suit the product.

Can I choose the model?

Yes. Send a model id from GET /catalog/models, custom_<id> for a model you made in Garmo Studio, or "auto" with a gender.

What does it cost?

One credit per image: count 4 makes four images and costs four credits. A failed image gives its credit back.

Other edits in the API

Get an API key

Test keys are free and need no pack.