Ideogram enable_copyright_detection: no such flag on Sume, what to do
Ideogram's precise edit takes an enable_copyright_detection flag. Sume's image request has no such field, so the check moves into your own pipeline.

Ideogram's Precise Edit reference (read 2026-10-06) lists enable_copyright_detection as an optional request parameter that is a copyright-detection toggle, beside quality, seed, num_images, async, webhook_url and dry_run. Sume's Image API request table has no field with that name or purpose.
Do not send it. Sume's own rule is that the catalog is the parameter allowlist, and a parameter the selected model does not list is rejected with 400 unsupported_parameter rather than dropped. That is better than silent loss, because your port fails on the first test call.
What a port should do
| Ideogram parameter | Sume request field | Port |
|---|---|---|
enable_copyright_detection | None | Add your own review step |
dry_run | None on /v1/images; the model endpoints list per-image pricing | Read the price line before you submit |
seed | Not served in v1 | Keep the first good output as the master |
async, webhook_url | mode: "async" or mode: "webhook" with webhook_url | Rename |
num_images | n | Read the model's range |
A review gate that fits a Sume job
- Store the prompt, the source URL and the job id with the output, so a reviewer can see what produced it.
- Hold anything that contains a logo, a character or a recognizable product look for a person to approve before it ships.
- Keep rejected outputs. They are your evidence if a brand team asks what was generated and why it was not used.
- If a job fails with
generation_rejected, correct the input; the docs say failed generations are not charged.
Why a rejection beats a silent drop
A port that quietly ignores a safety flag looks like it works and then ships unchecked images. A hard 400 on the first call tells you the control is gone and sends you to build it on purpose.
Write that decision down in the ticket for the migration. Name who reviews outputs, what they look for, and where the approval is stored.
Keep the check proportionate. A product photo with a neutral background needs a quick look. An image that carries a brand mark, a face or a character needs a named approver, and the approval should be saved with the job id so you can find it again later.
Sources
Related posts
More in Developers
- Ideogram API image URLs expire: save the edit, and what Sume returns
Ideogram's docs say image URLs expire and to download what you keep. Sume returns a hosted signed URL; here is a download step that works for both.
- Ideogram is_image_safe in the response: what Sume returns instead
Ideogram's precise edit returns an is_image_safe field per image. Sume has no such field: a refused input ends as a failed job with generation_rejected.
- Aspect ratios per Sume image model: Banana 2 has 14, Pro 10, GPT 8
Nano Banana 2 lists 14 aspect ratios on Sume, Nano Banana Pro 10 and GPT Image 2.5 8, each plus auto. The full lists, and how to read them from the API.
- image_size beats aspect_ratio on Sume /v1/images: send one, not both
On /v1/images, image_size has priority over aspect_ratio. Use exact pixels for GPT Image 2.5 and an aspect ratio plus resolution tier for Nano Banana.
Written by Sume