Is Higgsfield Genjutsu in your Sume catalog? Check, then price it
higgsfield-genjutsu appears in the catalog only when its provider is configured. List your models, then price a 4 to 30 second source at 480p or 720p.

Higgsfield Genjutsu (higgsfield-genjutsu) is Motion Transfer: one source video plus 1 to 8 reference images, at 480p or 720p, 4 to 30 seconds. The Video generation page says it is in the catalog only when its provider is configured, so list the models before you build on it.
Check the catalog
supported_durations and supported_resolutions come from the same response.
import os, requests
r = requests.get("https://api.sume.com/v1/videos/models",
headers={"Authorization": "Bearer " + os.environ["SUME_API_KEY"]})
r.raise_for_status()
m = {x["id"]: x for x in r.json()["data"]}
g = m.get("higgsfield-genjutsu")
print(g and (g["supported_resolutions"], g["supported_durations"][:1]))
print(sorted(m))Price by source length
Genjutsu preserves the source length, so you pay for the seconds of the clip you upload, rounded up. Send that length as duration; the docs require the inspected source length.
| Source length | 480p | 720p |
|---|---|---|
| 4 s | $1.59 | $3.41 |
| 8 s | $3.18 | $6.81 |
| 15 s | $5.96 | $12.77 |
| 30 s | $11.93 | $25.54 |
Request shape
Genjutsu takes a video_url and reference_image_urls on the Video Router wire. It accepts video references, not audio.
{
"model": "higgsfield-genjutsu",
"prompt": "Replace the dancer with the person in the photo",
"video_url": "https://example.com/dance.mp4",
"reference_image_urls": ["https://example.com/new-person.jpg"],
"resolution": "480p",
"duration": 8,
"mode": "async"
}Before you promise it in a product
Because availability depends on the provider being configured, code against the catalog, not against the id. Hide a Genjutsu option in your UI when the id is absent from GET /v1/videos/models, and show it when it appears.
Sume reserves the full price at submit, so the number in the table is also the amount that must be free in the balance when you send the request. If it is not, the call fails with 402 insufficient_credits before any provider work starts.
- Inspect the source clip first so the duration you send matches its length: Genjutsu preserves source length, 4 to 30 seconds.
- Reference images: 1 to 8 per request.
- Resolution: 480p or 720p; the Sume price at 720p is more than double 480p.
Sources
Related posts
More in Developers
- Is Vidu Q4 Preview on Sume? Check the live video catalog in Python
Sume's documented video catalog lists no Vidu model. A short Python check against GET /v1/videos/models confirms it before you build, plus the closest rows.
- Sume job returns 404 not_found? The key may not own that job
A valid job id can still return 404 not_found on the Sume API: API keys read only jobs their own member created. How to confirm it and what to change.
- A 10-minute render needs 11 to 12 jobs_wait calls, not one long one
On hosted MCP, jobs_wait holds at most 55 seconds, default 50. A ten-minute render is 11 or 12 waits on the same job id. Never resubmit the paid create.
- GPT Image 2.5 slow? OpenAI says jpeg is faster than png
OpenAI says jpeg is faster than png and complex prompts can take 2 minutes. How to set output_format on Sume and handle the 202 that follows.
Written by Sume