FLUX 3 Image images field: 2-10 references vs Sume's input_references

BFL's FLUX 3 Image images field takes 2 to 10 reference images, and Sume's input_references takes up to 10 on most models and 16 on ChatGPT Image 2.5. Order matters on both sides: BFL addresses references by index, and Sume's Ideogram 4.5 treats the first reference as the image to edit. FLUX 3 Image is not in the Sume image catalog today; check GET /v1/images/models for the live list. The catalog lists FLUX.2 Pro and FLUX.2 Flex.

Side by side

The BFL pages give the ranges and reference limits; the Sume numbers come from the Image API docs.

The single-image gap

BFL lists 2 as the minimum for the field, so a single-image edit may not follow the same path. The BFL pages cited here do not say what a one-image edit should use, so check BFL's docs for that case before you build.

The Sume request shape

On Sume, each reference is an object in an array. A model whose input_references descriptor is {min: 0, max: 0} is text-only and rejects references. Read the descriptor from the catalog rather than assuming.

{
  "model": "ideogram/ideogram-v4.5",
  "prompt": "same scene, warmer evening light",
  "input_references": [
    {"type": "image_url", "image_url": {"url": "https://example.com/source.jpg"}},
    {"type": "image_url", "image_url": {"url": "https://example.com/style.jpg"}}
  ]
}

Ideogram 4.5 order

For Ideogram 4.5 on Sume, the first reference is the source image and up to 4 more are references, 5 in total. An edit without aspect_ratio keeps the source shape.