Ideogram model pose variants API vs a Sume edit prompt

Ideogram's POST /v2/tool/model-pose-variants changes a fashion model's pose or camera angle from a plain-language instruction. Sume has no pose-specific endpoint: you call POST /v1/images with the source photo as a reference and describe the new pose in the prompt, up to 10 images per call with n.

Ideogram details are from its API reference; Sume details from Image generation, read 2026-10-01.

What does the Ideogram tool take?

A multipart request with a required source_image (max 50MB) and a required instruction, for example a front-facing catalog stance or a right-facing walking profile. The reference says the model, garment, styling and scene are preserved. aspect_ratio is optional and, when omitted, derived from the source image.

How does that map to Sume parameters?

How do I ask for several poses at once?

Put the source photo in input_references, write the pose in prompt, set aspect_ratio: "auto" and raise n. The docs say to request up to 10 images per call with n but to read the n range descriptor from the catalog, because per-model ceilings are lower. Keep the garment and scene instructions explicit in the prompt, since Sume does not advertise a preservation guarantee the way the Ideogram tool describes one.

For garment-only product shots see AI ghost mannequin.

Will an unsupported parameter be ignored?

No. A request that sets a parameter the selected model does not list is rejected. Check supported_parameters for the model before you send fields copied from another API.