Skip to main content

Product use cases · Screen production · Images and video

One product shot, every size the media plan asks for.

Start with a cobalt-blue perfume bottle and a simple white label. Plan a 1:1 packshot, a 9:16 story image, and a five-second camera push-in from a selected still.

Best for
Performance marketing teams, agencies and commerce brands producing size and message variants at volume.
What you get
A variant matrix where every cell points back to one committed source file, one endpoint and one price you saw before the run.

Concept illustration

A creative brief to try

Turn one perfume bottle into three assets

  1. 01

    Use a sharp bottle reference. Make the square packshot first with a clean background and enough space around the cap.

  2. 02

    Return to the same bottle reference for the vertical story. Change the composition and background, keeping the bottle shape and label placement.

  3. 03

    Animate the selected still with a slow five-second push-in. Keep the bottle still and add campaign text in your editor.

Before you keep the result

Inspect the bottle silhouette, cap, label, and reflections. Watch for changing letters or bending glass throughout the clip.

Start with your own reference images, follow the scene plan and review each result.

Open this scene in the Playground →

Recommended models

191 matching models

Matches use the tasks and required capabilities published in the catalogue. Availability and current prices come from each model page.

Image editing
56
from $0.0018 / image
GPT Image 2.5 Sunburst · Image Editing →
Text to image
61
from $0.0018 / image
GPT Image 2.5 Sunburst · Text to Image →
Image to video
56
from $0.011 / output second
Seedance 1.5 Pro · Image to Video →
Reference to video
18
from $0.011 / output second
Seedance 2.0 Mini · Reference to Video →

What one run costs

Worked out from the live catalogue, at the settings named under each figure.

GPT Image 2.5 Sunburst · Image Editing
$0.0018
1 × $0.0018 · image · lowest tier
P-Image LoRA · Text to Image
$0.006
1 × $0.006 · image
Seedance 1.5 Pro · Image to Video
$0.056
5 × $0.0112 · output seconds · lowest tier

Catalog estimates refresh hourly. Check the estimate for your settings before running; final charges follow the model’s usage and billing policy.

How to build it

Save the original product photo and build a matrix of background, framing and motion choices. Estimate each variant with its own settings before submission; one quote applies to one request. Review the bottle, label and cap in every result, then create a revised variation only for the cells that need it.

  1. 01

    Ask for an upload ticket

    POST/api/v1/common/upload-url

    The ticket is valid only for the byte count you declared: JPEG, PNG, WebP and GIF up to 10 MiB each. Media that already has a public HTTPS URL skips this step entirely.

  2. 02

    Commit the uploaded file

    POST/api/v1/files/{fileId}/commit

    The commit checks the bytes and hands back a spicy:// reference, which is what a model's media fields accept. Committing the same file again changes nothing.

  3. 03

    Price the run before it exists

    POST/api/v1/jobs/quote

    The same validation and admission rules as creating the task, except nothing is created and no money is held. The signed quote lasts five minutes.

  4. 04

    Create the task

    POST/api/v1/jobs/createTask

    202 means accepted, not finished. Send an Idempotency-Key, and if the price must not move, send the jobs/quote result together with the amount you agreed to.

  5. 05

    Re-run one failure, not the batch

    POST/api/v1/jobs/retry

    A retry creates a new task from a failed or expired one and leaves the source untouched. Schema, price, permissions, balance and availability are all evaluated again, so a retry runs on today's terms.

  6. 06

    Move the file into your own storage

    POST/api/v1/common/download-url

    Mint a short-lived download URL for the output and copy it into your own bucket within the same job. Results stay on our CDN for 14 days.

Copy a request

This example starts with the scene described on this page. Fields come from the selected endpoint’s schema. Replace any example.com reference URL with your own image or committed file URI before sending.

GPT Image 2.5 Sunburst · Image Editing →

Where this usually snags

Platform behaviour that is cheaper to design around than to discover in production.

  • A quote holds the price for five minutes

    jobs/quote runs the same validation and admission checks as creating the task, then signs a price that is good for five minutes. Quote when you are about to submit, not when you are planning; an expired quote means the run prices itself again.

  • Aspect ratio is a list, not a number

    Resolution, aspect ratio and length come from each endpoint's own input schema as a closed set, and extra properties are refused. Check that the frame you ship in is on that list before you pin an endpoint to a format.

  • Reference fields need a model that takes references

    Input is validated against that endpoint's own schema and extra properties are refused, so a reference sent to a model without the capability is a 400, not a field quietly ignored.

Launch checklist

  1. 01Use a sharp bottle reference. Make the square packshot first with a clean background and enough space around the cap.
  2. 02Save the original product photo and build a matrix of background, framing and motion choices. Estimate each variant with its own settings before submission; one quote applies to one request. Review the bottle, label and cap in every result, then create a revised variation only for the cells that need it.
  3. 03Read each endpoint's frame options before assuming a size exists; aspect ratio is a closed enum per endpoint, not a free number.
  4. 04For a completed result that needs a visual revision, create a new task with revised inputs. Retry is for failed or expired tasks and uses current schema, pricing and availability. Keep approved pieces and replace only the affected variants.

Models that match this workflow

Explore model families for this workflow. Open a model page to review its inputs, pricing and current availability.

Frequently asked questions

191 published models match the listed tasks and required capabilities. Compare their inputs, examples, availability and current prices before choosing one.

About this workflow

Save the original product photo and build a matrix of background, framing and motion choices. Estimate each variant with its own settings before submission; one quote applies to one request. Review the bottle, label and cap in every result, then create a revised variation only for the cells that need it. Failed or timed-out tasks are usually not charged. Some models charge for content refusals as stated on their model page; check those billing terms before submitting.