Gym class schedule promo: a silent Wan clip and three timed cues
A gym class schedule video on Sume: one 6-second Wan 3.0 480p clip and one caption job with three timed cues. $0.575, no speech-to-text, read 2026-10-10.

A gym schedule promo costs $0.575 on Sume: one 6-second Wan 3.0 clip at 480p ($0.3750) and one caption job ($0.20) that prints three timed class lines over it. The caption job accepts cues, so it needs no speech and works on a silent clip.
Schedules are text, and video models are bad at text. The workable split is a video with no words for the mood, and overlay captions for the times. If the timetable changes next month, you recaption the same clip for $0.20 instead of regenerating. Rates come from the Sume pricing code (read 2026-10-10).
What is the endpoint chain?
Two writes, no Timeline, because the clip already has the right length.
POST /v1/videoswithwan-3.0,duration: 6,resolution: 480p,aspect_ratio: 9:16and a prompt for a bright, empty studio with soft motion.POST /v1/video-captionswithvideo_urlandcues: 'Mon 6:30 Spin' from 0 to 2, 'Wed 7:00 Yoga' from 2 to 4, 'Sat 9:00 Bootcamp' from 4 to 6.- Send only
cues; do not mix it withscript_text,wordsorsegments. - Use
Idempotency-Key: schedule-2026-11so a rerun returns the original job.
What does it cost?
Resolution is the only choice that changes the bill; the caption job is fixed.
| Resolution | Clip | Captions | Total |
|---|---|---|---|
| 480p | $0.375 | $0.20 | $0.575 |
| 720p | $0.750 | $0.20 | $0.950 |
| 1080p | $1.500 | $0.20 | $1.700 |
What are the limits?
Cues need text, start and end in seconds. Three lines in six seconds is the readable maximum; a longer schedule needs a longer clip or more cues over a second clip. Wan accepts 2 to 30 seconds, and 30 seconds at 480p is $1.875.
The picture should have no people performing a specific exercise, because models can render odd poses. Use an empty studio, equipment or hands on a mat. Keep claims such as results or health benefits out of the cues; they are your responsibility, not the model's. The Video captions page lists styles and design overrides.
A silent clip with captions has no sound at all. If you want a bed, put it through Timeline with a free BGM track, which adds a $0.10 render.
How do I handle schedule changes?
Keep the clean clip and reburn the cues. A monthly schedule is then $0.20 a month after the first $0.375 clip. For a larger set of weekly posts, see the 12-item queue, and for adding sound see a music bed.
Pre-publish checklist
View the result on a phone at arm's length. Each cue should be readable in two seconds, so keep the text short and the times in the same format.
Check the times against the real timetable on the day you post.
- Three cues for a six-second clip.
- One style for the whole month.
- Keep the clean clip for next month's recaption.
Sources
Related posts
More in Use cases
- Holiday-market walking tour teaser: one 30-second clip for $1.88
A local guide can post a 30-second holiday market tour teaser from Wan 3.0 for $1.88 at 480p. Where 30 seconds fits on Google Business and Shorts.
- India Rule 3(3): metadata, identifier and no-removal for AI media
India's IT Rules now require tools that make synthetic media to label it and embed permanent metadata. What the rule text says and what to check on a Sume file.
- Instagram Align for Reels: match AI shots with the last frame
Instagram's Align lines up objects between recorded clips. For AI shots, take the previous clip's last still and send it as the next clip's first frame.
- Japan's AI display duty for election video starts 1 March 2027
Japan's election-law bill adds Article 142-5: AI-made or altered campaign images and video must say so on screen. Dates, exceptions, and a burn-in step on Sume.
Written by Sume