Genjutsu Object Swap API: Sume serves Motion Transfer only
Higgsfield pairs Motion Transfer with Object Swap in Genjutsu. Sume's Genjutsu row serves Motion Transfer only; Object Swap is rejected by name.
There is no Object Swap through Sume. Higgsfield's Genjutsu page lists two modes, Motion Transfer and Object Swap, but Sume's Genjutsu catalog row serves Motion Transfer only, and its public description rejects Object Swap by name.
Higgsfield's modes are from its Genjutsu page, read 2026-09-30. Sume's side is from the catalog in code and the Video Router docs.
What are the two Genjutsu modes?
Per Higgsfield, Motion Transfer keeps the motion, camera and timing and rebuilds everything else from your references. Object Swap modifies selected elements, such as characters, outfits, products or locations, while preserving the original shot's composition.
What does Sume's row serve?
| Mode | Higgsfield page | Sume catalog row |
|---|---|---|
| Motion Transfer | Listed | Served: one source video plus reference images |
| Object Swap | Listed | Not served by this model |
What can I use for an object-level edit instead?
Sume's catalog documents video_to_video edits for the Gemini Omni Flash 1.1 row: send video_url and a prompt that describes the edit, for example replacing one object and keeping everything else. That is a different model with its own limits, not Object Swap, and its output may differ. See Gemini Omni video edit prompts.
Sources
Related posts
More in Models
- GPT Image 2.5 cached input price: does Sume pass it on?
OpenAI lists cached input at $1.25 text and $2 image per million tokens for Sunburst. Sume's docs give estimates and reserved cost, not a cache discount.
- GPT Image 2 supported aspect ratios on Sume: 5:4, 9:8, 4:5
On Sume, gpt-image-2 accepts 5:4, 9:8 and 4:5 on top of the common ratios, plus custom pixels with a 3:1 limit. The exact list and the 9:8 size to send.
- Grok Imagine image edit limit: xAI 5 sources, Sume catalog
xAI's Aug 28 release notes raise image editing to 5 source images. On Sume, Grok is edit-capable with n of 1; read input_references per model.
- grok-imagine-image-quality retires Nov 2: what about quality?
xAI retires grok-imagine-image-quality on November 2, 2026 and serves it as grok-imagine-image-2.0 at low quality. Sume's x-ai/grok-image takes no quality.
Written by Sume