FLUX 3 Image 10 reference images vs Sume's 1-10 image_urls
FLUX 3 Image takes up to ten references named by prompt position. Sume's image edit takes one to ten public image_urls, with no FLUX 3 id in the docs.

Yes, FLUX 3 Image takes up to ten reference images, and Sume's image edit also takes one to ten. On Sume the field is image_urls, 1 to 10 public HTTPS URLs, on the Image 1.0 compatibility route. Sume's docs list no FLUX 3 model id, so this is a comparison of limits, not a way to call FLUX 3.
Black Forest Labs' side is from its release notes and Sume's from the Image 1.0 and Image generation docs, all read 2026-10-01.
How does FLUX 3 Image use ten references?
The release notes say to name each image by its position in the prompt, to restyle an image or combine references into a new composition. There is one endpoint and no mode field: the prompt decides whether to generate or edit.
How many references does Sume accept?
It depends on the route. Image 1.0 is a compatibility alias for Image Router Auto and is retiring soon; new integrations are told to use POST /v1/images with model: "sume/auto". Per-model reference ceilings on the Image Router are in each catalog record, and a model with {"min": 0, "max": 0} is text-to-image only.
| Source | Field | Limit |
|---|---|---|
| BFL release notes | Prompt positions | Up to ten references |
| Sume Image 1.0 route | image_urls | 1 to 10 public HTTPS URLs |
| Sume Image Router example | input_references | Range descriptor, example max of 10; read it per model |
How do I send ten references on Sume?
Put the URLs in image_urls and describe each one in the prompt. Sume's docs do not define a positional naming scheme for the Image 1.0 route, so refer to images in plain words ("the first image") and check the result.
curl -X POST https://api.sume.com/v1/image-1.0/generate \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: image-refs-001" \
-d '{
"prompt": "Combine the product from the first image with the room from the second",
"image_urls": [
"https://example.com/product.png",
"https://example.com/room.jpg"
],
"quality": "medium"
}'What else can I add to an edit?
A mask_image_url alongside image_urls makes a masked edit; see mask_image_url vs mask_url. For what Sume's catalog does serve from FLUX, read FLUX 3 Image vs FLUX.2.
Sources
Related posts
More in Models
- FLUX 3 x mimic: one backbone for video, audio and robot actions
BFL says FLUX 3 is one model trained jointly on images, video and audio, and FLUX-mimic adds robot actions. What that does and does not tell a media API user.
- FLUX 3 video continuation caps at 15 s: extending a clip on Sume
BFL cut FLUX 3 video continuation (v2v) to 15 seconds on Aug 17; other modes stay at 20. Sume has no continuation mode: chain a last frame instead.
- FLUX 3 Video 4K uhd (3840x2176) and which Sume video models do 4K
FLUX 3 Video's uhd resolution is 3840 x 2176 at 16:9. On Sume, a 4K request goes to a model whose supported_resolutions lists it; here is what the docs say.
- Flux TTS expressivity -2 to 2 vs Sume's emotion guide
Deepgram's Flux TTS expressivity runs -2 to 2 (0 nominal). Sume TTS has no such dial: generation_config takes volume, speed and a free-text emotion guide.
Written by Sume