Next-episode end card in Timeline 1.0: held still, fade, cliffhanger
Add a next-episode end card to a micro-drama with a Timeline 1.0 still slot: stills hold static, fades run up to 1 s, render is $0.10 per output minute.

How do you add a 'next episode' end card to an AI micro-drama? Put a still image in the last Timeline 1.0 video slot, hold it for a second or two with a fade transition, and render. Stills are static holds, a transition is capped at 1 second, and the render costs $0.10 per ceil output minute.
Serialized vertical drama lives on the last frame: each episode ends on a cliffhanger, and YouTube's new Shorts series feed puts the next episode one swipe away. A held card gives the viewer a beat to tap.
The request
Timeline 1.0 takes one audio spine plus ordered video slots. Every URL must be a media.sume.com artifact or asset of your workspace; import files first. Idempotency-Key is required. The last slot below is a still, held for 2 seconds after a 0.5 second fade.
{
"audio": { "url": "https://media.sume.com/artifacts/artf_demo/ep3-voice.wav", "duration_seconds": 45 },
"video": [
{ "source_url": "https://media.sume.com/artifacts/artf_demo/ep3.mp4", "start": 0, "duration": 43 },
{ "source_url": "https://media.sume.com/artifacts/artf_demo/next-card.png", "start": 43, "duration": 2,
"transition": { "type": "fade", "duration": 0.5 } }
]
}Rules that matter here
video[0].start must be 0, later starts must increase, and each slot's duration must be at least 0.2 seconds. Coverage can stop at most 0.5 seconds before the end of the spine, so make the card's end match the voice track length. A transition is limited to 1 second and 50% of the shorter neighbor, and it cannot sit on the first slot (transition_on_first_segment).
If you give the still a motion field, the job accepts it and ignores it, reporting motion_ignored; that is a soft warning, not a failure.
| Item | Limit |
|---|---|
| Still slot | Static hold |
| Slot duration | 0.2 s minimum |
| Transition duration | 1 s maximum, 50% of shorter neighbor |
| Output | 1080x1920 MP4 by default |
| Price | $0.10 per ceil output minute |
Check before paying
POST /v1/timeline-1.0/plan is an unbilled compile preflight. It returns duration_seconds, segment_count, billable_minutes and an estimated cost, and it does not create a job. Run it in CI for every episode and you catch a bad start time before a render slot is spent.
Shorts series episodes are portrait or square and up to 180 seconds, so a 45-second episode with a 2-second card fits with room to spare.
Reusing the card
The end card is a template. Keep the still, the fade and the duration in one timeline fragment, then swap only the episode number and the next-release text each week. Because the plan endpoint is free, you can test each variant without paying. The only billed step is the final render.
Sources
Related posts
More in Use cases
- 90 hook drafts at 360p, then 6 finals: a 30-product hook test
Thirty products times three hooks is 90 Omni drafts at 360p for $10.125; six winners at 1080p add $9.00. The whole test costs $19.125 on Sume.
- One 58-second vertical master that fits Shorts, Reels, and TikTok ads
A 58-second 1080x1920 master fits YouTube Shorts, the Reels 3-minute line, TikTok reservation and TopView, and Google's Shorts-ad guide. Render it and verify.
- One ad, five cities: swap the backdrop per market with Omni edit
Localize a street-scene ad by editing the backdrop per market with Gemini Omni Flash. Prompt pattern, five 8-second edits at 720p for $5.00, and what to verify.
- One AI image a day for a year: 365 images cost $1.825 to $96.27
A daily image for 365 days costs $1.825 on Soul at 720p, $9.13 on Grok, $13.69 on Flux 2 Pro and $96.27 on ChatGPT Image 2. Table and a scheduling note.
Written by Sume