Recraft V4 Styles: 10 references, and the Sume Recraft id
Recraft V4 Styles takes one to 10 reference images in Precise or Flexible mode. Sume lists recraft/recraft-v4 as text-to-image only and rejects references.

Recraft V4 Styles, released in August 2026, matches a style from one reference image or as many as 10, in a Precise or a Flexible mode. Sume does not list a V4 Styles id. Its Recraft entry is recraft/recraft-v4, a text-to-image model, so a request with input_references for it is rejected.
Recraft facts are from its V4 Styles page; Sume facts from Image generation and the catalog code, read 2026-10-01.
What does Recraft V4 Styles do?
The page says you attach a reference, generate, and nothing needs training or fine-tuning. Similar references sharpen the match, and a wider mix widens the range. References can come from other generators, brand libraries or old campaign assets. It is available in Recraft Studio and through the Recraft API.
What is the difference between Precise and Flexible?
Precise is for "match this exactly" and follows a reference's texture, palette and rendering. Flexible is the default and the looser read, for the mood of a reference rather than a copy. The page names four models: recraftv4_styles, recraftv4_styles_pro, recraftv4_styles_vector and recraftv4_styles_pro_vector.
| Item | Recraft V4 Styles page | Sume |
|---|---|---|
| Model ids | Four recraftv4_styles* ids | recraft/recraft-v4 only |
| Reference images | One to 10 | Not accepted on recraft/recraft-v4 |
| Precise / Flexible | Two modes | No equivalent parameter |
| Output | Not stated on the page | webp for this entry |
Does the Sume Recraft id take references?
No. The catalog row for recraft-v4 is marked text-to-image only, with image_urls not accepted. The Sume input_references field is documented as "Reference images for image-to-image", but the docs say a parameter the selected model does not list is rejected with 400 unsupported_parameter rather than silently dropped. A Precise or Flexible setting has no field to carry it either.
How can I do style-matched work on Sume?
Pick a model that lists reference input in the catalog instead of sending references to Recraft. Reference images in the image API covers which ones do, and which image model compares the catalog. Check the model's listed parameters before you build a flow around it.
Sources
Related posts
More in Models
- Runway Compositing Nodes vs Sume Timeline compose
Runway added compositing nodes to blend images and video in Workflows. Sume's Timeline compose puts one still and one video in one MP4: stack or overlay.
- Seedance 2.5 480p: ratio 864:496 gives 854x480, Sume uses resolution
On Runway, Seedance 2.5 ratios 864:496 and 496:864 return 854x480 and 480x854. On Sume you ask for 480p with resolution plus aspect_ratio, never pixels.
- Runway video_to_hdr alpha and ProRes 4444: what Sume returns
Runway's /v1/video_to_hdr now keeps source alpha and delivers ProRes 4444. Sume docs describe no alpha or ProRes output; a completed video job is a download.
- Seedream 5.0 Flash: which Seedream 5 id Sume's catalog lists
Higgsfield added Seedream 5.0 Flash on Sep 23, 2026. Sume's image catalog lists seedream-5-lite and no Flash id. Check GET /v1/images/models before calling.
Written by Sume