Which Sume image models take no output_format? Soul and Ideogram 4.5

Two Sume image models list an empty output_format enum: Soul (higgsfield/soul) and Ideogram 4.5 (ideogram/ideogram-v4.5), read 2026-10-03. The provider picks the file type for both, so you leave output_format out of the request. A third row, Recraft V4, lists webp as its only value. The other sixteen rows (both ChatGPT Image 2.5 ids included) list png, jpeg and webp.

The Image API page says a request that sets a parameter the selected model does not list is rejected with 400 unsupported_parameter rather than silently dropped. For a model whose descriptor is an empty list there is no valid value to send.

What each group accepts

Count taken from the catalog descriptors.

Read the descriptor instead of hard-coding

A client that sets output_format by default breaks on the empty-list models. The safer pattern is to ask the catalog and send the field only when its list is not empty and contains your choice.

import os
import requests

headers = {"Authorization": f"Bearer {os.environ['SUME_API_KEY']}"}
models = requests.get("https://api.sume.com/v1/images/models", headers=headers, timeout=30).json()["data"]
wanted = "webp"
for m in models:
    allowed = m["supported_parameters"]["output_format"]["values"]
    ok = "send it" if wanted in allowed else "omit it"
    print(m["id"], ok)

What about compression and transparency

output_compression is in the schema but no model advertises it, so it returns 400 for everyone. Transparent stills come from background: "transparent" on ChatGPT Image 2.5, not from picking a format. If you need a specific file type from Soul or Ideogram 4.5, convert after download, for example with Pillow.