Employee onboarding videos: ten 30-second Wan 3.0 modules for $37.50
Ten 30 second onboarding clips on Sume's Wan 3.0 cost $37.50 at 720p, $18.75 at 480p or $75.00 at 1080p. Concurrency by plan and what to keep out of frames.

Ten 30 second onboarding videos cost $37.50 at 720p on Sume's wan-3.0: 300 seconds at $0.125 a second. At 480p it is $18.75, and at 1080p it is $75.00.
The math
The rates are provider list ($0.05, $0.10, $0.20 a second) times Sume's 1.25, from the repository pricing tables. Duration is limited to 2 to 30 seconds per request, so a 30 second module is a single clip, with no stitching.
| Resolution | Per module | Ten modules | Twenty modules |
|---|---|---|---|
| 480p | $1.875 | $18.75 | $37.50 |
| 720p | $3.75 | $37.50 | $75.00 |
| 1080p | $7.50 | $75.00 | $150.00 |
Concurrency by plan
Ten jobs do not all start at once on every plan. Sume accepts valid jobs as queued and starts them under a plan limit of 1 (Free), 4 (Pro), 8 (Startup) or 20 (Scale) jobs processing, with queue capacity of 5, 20, 40 and 100 (generation admission docs, read 2026-10-05). On Pro, ten jobs run as four, four, two; on Startup as eight and two; on Scale, all ten together. Free accepts six jobs in total (1 processing plus 5 queued), so submit ten in two waves.
What goes in the ten modules
Ten modules is a plan, not a minimum. Common candidates are a welcome, an org map, a tools tour, a security reminder, a time-off explainer, an expenses explainer, a meeting norms recap, a benefits overview, a support contacts card and a first-week checklist. Each fits in one 30 second clip if the script is one idea long.
Keep real policy text out of generated frames. A model can draw a calm office and an arrow pointing at a laptop; it should not be the source of truth for a policy number. Put exact policy words in narration or a caption layer that you review, and use the generated clip as the visual.
Consistency across ten clips is the other cost. Reuse the same style sentence in every prompt, and the same reference images if you have a brand look. Sume allows up to 10 reference images on a Wan 3.0 request, so a small set of brand stills can ride along with each module.
Where a person speaks
If a module needs a person to speak on camera, a video model is the wrong tool; Sume's docs say video models do not lip-sync to generated narration, and talking clips go through Avatar Video instead. Use Wan 3.0 for visuals and put voice-over in your editor.
A final planning note: the $37.50 figure assumes no retries. Add one re-render per three modules (about 3 x $3.75 = $11.25) and the plan is $48.75.
Submit the ten
Submit in a loop, store each job id, and poll every job to a terminal state. This script submits the modules and prints the ids; it takes the prompts from a list and uses one idempotency key per module.
import asyncio, json, os, urllib.request
MODULES = ["Welcome to the team", "How we meet", "Where tools live"] # ten in practice
def submit(n, topic):
body = {"model": "wan-3.0", "resolution": "720p", "duration": 30,
"aspect_ratio": "16:9",
"prompt": f"Calm office scene, no dialogue: {topic}. Soft light, wide shots."}
req = urllib.request.Request("https://api.sume.com/v1/videos",
data=json.dumps(body).encode(), method="POST",
headers={"Authorization": "Bearer " + os.environ["SUME_API_KEY"],
"Content-Type": "application/json",
"Idempotency-Key": f"onboarding-{n:02d}-v1"})
return json.load(urllib.request.urlopen(req))["id"]
async def main():
jobs = await asyncio.gather(*(asyncio.to_thread(submit, i, t) for i, t in enumerate(MODULES, 1)))
print(jobs)
asyncio.run(main())Balance and failures
A wallet check comes before the batch. Sume reserves the billable amount at submit, so a workspace needs the balance for the jobs it submits at once. Ten 720p modules reserve $3.75 each, which is $37.50 in total if all are accepted together. A submit without enough balance fails with 402 insufficient_credits before provider work starts, per the generation admission docs.
When a job fails, Sume's docs describe a public error and a terminal state rather than a silent hang, so read the failure and resubmit only that module with a new idempotency key.
Keeping modules consistent
Ten modules made by ten separate prompts drift in look. Write one shared style paragraph (setting, color, camera, pace) and paste it at the top of every prompt, then add only the topic sentence. That makes the set feel like a series and keeps edits cheap, because changing the style changes one paragraph.
Budget one spare run per module if your approvals are strict. Ten extra 30 second runs at 720p add $37.50, which doubles the plan, so do the first pass at 480p ($1.875 per module) and approve the wording before paying for 720p.
One key per module
Submit each module with its own Idempotency-Key, such as onboarding-01-v1. A retry after a network error then returns the original job and does not bill a second one. Keep a ledger of module number to job id.
Sources
Related posts
More in Use cases
- English podcast clip to a Spanish short: dub and captions, $0.38
An English podcast clip turned into a Spanish short costs up to $0.38 on Sume: trim, detach, transcribe, speak, render and one caption job at $0.20.
- Engraved and monogrammed gifts: one order-by cue per lead time
Personalised gifts need earlier cutoffs than stock items. Compute each order-by date from production plus transit days and burn all lines in one caption job.
- Place the EU AI icon where app UI and captions will not cover it
The EU icon page says no overlay should obstruct the icon on video. TikTok publishes ad safe-zone files; use compose layout and caption placement to stay clear.
- EU AI icon on video: burn it in, do not rely on a platform badge
EU icon guidance says the label must be embedded in the video and kept on reshare or download. A burned-in overlay does that; a platform badge does not.
Written by Sume