Avatar video is 720p only: fit it into a 1080x1920 series
Sume Avatar Video 1.0 renders at 720p. Timeline 1.0 outputs 1080x1920 by default and fits a slot with cover or contain, at $0.10 per output minute.
Avatar Video 1.0 renders only at 720p, so to deliver a 1080x1920 series put the avatar clip into a Timeline 1.0 slot. Timeline outputs 1080x1920 by default and a slot's fit decides how a 720p source fills the frame. The render fee is $0.10 per output minute, rounded up.
The numbers
None of this changes the avatar itself. The avatar is created once, and each talking video is generated from it at 720p, with the 4 to 60 second duration window and three quality tiers. Timeline is a separate step that only arranges the finished files.
| Item | Value |
|---|---|
| Avatar resolution | 720p only |
| Avatar aspect ratios | 1:1, 3:4, 9:16 (default), 4:3, 16:9 |
| Avatar duration | 4 to 60 seconds |
| Timeline default output | 1080 x 1920 |
| Timeline fee | $0.10 per ceil output minute |
| Slot options | source_url, start, duration, source_in, fit, transition |
What 720p to 1080p does
Timeline scales the source to the output frame; it does not add detail. A 9:16 avatar clip at 720p is 720x1280, and filling a 1080x1920 frame with it is an upscale by 1.5. On a phone that looks fine for a talking head, and on a big screen it looks softer than the 1080p b-roll you may cut around it. If that matters, mix the avatar with shorter close shots and put the presenter in a smaller part of the frame.
A minimal body
This puts one avatar clip on the timeline with its own audio as the voice track. Replace the URLs with your hosted files. Run /v1/timeline-1.0/plan first; it is unbilled and returns the problems before you pay.
Timeline refuses some inputs, which is why the plan step exists: the first video slot must start at 0, there can be up to 200 video slots and up to 20 audio parts, and more than 8 adjacent fades are refused. A one-slot episode is far inside every limit, so the plan is a cheap check that your URLs are reachable.
{
"audio": {
"duration_seconds": 30,
"parts": [
{ "source_url": "https://media.sume.com/your-avatar.mp4", "start": 0, "duration": 30 }
]
},
"video": [
{
"source_url": "https://media.sume.com/your-avatar.mp4",
"start": 0,
"duration": 30,
"fit": "cover"
}
]
}Check the field names
The body above follows the shape we read in the Timeline docs, but the audio and video fields have more options than this example uses (transitions, soundtrack with duck_db, up to 20 audio parts and up to 200 video slots). Confirm names on the Timeline page and send it to /plan before the first paid render.
A rule of thumb for the fee: a 30-second episode rounds up to one output minute, so the Timeline step is $0.10 on top of the avatar video. A 40-episode season of such files is $4.00 of Timeline fees, which is small next to the avatar seconds themselves.
Sources
Related posts
More in Sume Avatar 1.0
- Avatar quality for a 30-second UGC ad: standard, plus or max
A 30-second avatar UGC ad with a product image costs $5.82 at standard, $7.74 at plus and $17.40 at max. When each tier is worth it.
- Avatar video quality tiers: test one script at all three
How to choose between standard, plus and max in Sume Avatar 1.0: render one script three times, compare on a checklist, and keep the cheapest tier that passes.
- Cameo in an AI episode: Avatar Face Swap (Beta), 4 to 15 s source
Sume's Avatar Face Swap (Beta) puts your avatar into a source video of 4 to 15 seconds and requires a quality setting. Use it for a cameo, not a whole episode.
- Griffin-style follow-ups with rendered avatar clips and branching
Griffin-Lite reacts live. Until you can use it, approximate a guided conversation with a set of pre-rendered Sume avatar clips and your own branching logic.
Written by Sume