Seasonal avatar host from a prompt, profile or photo for Christmas
Create a reusable Sume Avatar 1.0 handle from a prompt, structured profile props or a reference photo, then reuse it across holiday clips.
Three inputs, one handle
To get a host you can reuse for a holiday campaign, call POST /v1/avatar-1.0/generate with an avatar_handle and one input: a prompt, a profile (props), or a reference image (photo). Poll the job, then use the returned handle in every later talking-video request.
The handle may include a leading @; Sume stores it normalized without the @. Make one handle per campaign host and reuse it, so every clip from Black Friday to Christmas shows the same person.
Profile and photo bodies
Use props when your app already has structured traits (ethnicity, sex and age appear in the docs example), and photo for a reference image that must be a fetchable public HTTPS image URL.
curl -sS -X POST https://api.sume.com/v1/avatar-1.0/generate \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: holiday-host-001" \
-d '{
"avatar_handle": "holiday_host",
"input": { "type": "props", "ethnicity": "Asian", "sex": "female", "age": 28 }
}'Which input to pick
| Input | API type | Use when |
|---|---|---|
| Prompt | prompt text | You want an avatar from text only |
| Profile | props | Your app already holds structured traits |
| Image | photo | You have a reference image |
Use the handle
Every request creates a job. When it is completed, read the avatar with GET /v1/avatar-1.0/avatars/{id}, then pass avatar_handle to the talking-video route or to Face swap. Use a photo of a real person only with their consent, and keep that consent on file with your campaign records.
Sources
Related posts
More in Sume Avatar 1.0
- How to make an AI UGC ad look less staged with Avatar 1.0
Less-staged AI UGC comes from the first frame: phone-style framing, a casual scene prompt, an approved preview, then the final render. The levers Sume exposes.
- MCP avatar-image-to-video_create: choose Fabric or H3 Max
On Sume's hosted MCP, one tool, avatar-image-to-video_create, makes still-plus-audio talking clips. A model field picks Fabric or MiniMax H3 Max lip sync.
- Recurring AI lead for a TikTok series: $0.95 once, then per episode
Make a series lead once with Sume Avatar 1.0 for $0.95, then bill each episode by the second. 60, 180 and 300 second costs at three quality tiers.
- Shortest AI avatar video: 4 seconds, from $0.74 on Sume Standard
Sume avatar videos run 4 to 60 seconds. Per-second rates for Standard, Plus and Max, the 4 second floor, and what 15, 30 and 60 second clips cost.
Written by Sume