Name your avatars: a handle scheme that fits 2 to 30 characters
Sume avatar handles allow letters, digits, periods and underscores, 2 to 30 characters. Build a team naming scheme that passes validation and stays readable.
A Sume avatar handle must be 2 to 30 characters of letters, digits, periods and underscores. It cannot start or end with a period or underscore, and it cannot contain two of them in a row. Hyphens are not allowed. Pick a scheme that fits these rules before the first avatar exists, because the handle is what every later talking-video request uses.
The rules, in one table
These come from the Avatar 1.0 request schema in the published OpenAPI and the create-avatar docs.
| Rule | Detail | Example that fails |
|---|---|---|
| Length | 2 to 30 characters, not counting a leading @ | a |
| Characters | Letters, digits, underscores, periods | product-host |
| Edges | No leading or trailing period or underscore | _host or host. |
| Repeats | No consecutive periods or underscores | host__one |
| Case and @ | A leading @ is allowed and stripped; stored lowercase without @ | Not a failure, just normalized |
A scheme that survives the rules
Because hyphens are out, use underscores between parts and periods only when you want a visible group break. A readable pattern is brand, role and version: acme_host_v2. It is 12 characters, which leaves room. Avoid putting a customer or campaign name in a handle, since handles are meant to be reused across many videos.
- Brand or team first, so a sorted list groups avatars together.
- Role second: host, demo, support, founder.
- Version last, so a re-created face does not overwrite the old handle.
- Keep a spreadsheet or table of handle, source, owner and creation date.
Why the lowercase detail matters
Sume lowercases the handle when it stores it. If your own system keeps Acme_Host and compares it with the stored acme_host, a case-sensitive check will say they differ. Lowercase handles in your code at the moment you write them to your database.
Read them back
List the avatars in a workspace with GET /v1/avatar-1.0/avatars, or read one by id with GET /v1/avatar-1.0/avatars/{id}. A handle is only useful for a talking video once the avatar is ready, so check its state first rather than assuming that a returned job id means a usable avatar.
In a multi-scene video, each scene in video_inputs can carry its own avatar reference, and the request can hold up to 20 scenes. Today one final video resolves to one avatar, so plan one handle per video.
Retiring a handle
Do not reuse a handle for a different face. Create a new one with a version bump and move your templates to it. The old handle stays in your history so old videos still make sense.
Sources
Related posts
More in Sume Avatar 1.0
- Let users pick an avatar in your app with GET /v1/avatar-1.0/avatars
Build an avatar picker on the Sume Avatar 1.0 list route: read ready avatars server-side, cache them, and pass the chosen handle to talking-video.
- Rate avatar clips like Griffin-Lite: naturalness, trust, flow
Tavus reports Griffin-Lite averages of 5.4 naturalness, 5.6 trust and 4.9 conversation flow. Score your own avatar clips on the first two; a test costs $14.69.
- Sume Avatar max costs 2.24x plus and 3x standard: when to pay it
Sume Avatar 1.0 is $0.184 a second on standard, $0.245 plus, $0.55 max: ratios 1 : 1.33 : 2.99. Iterate on standard, ship plus, pay max for hero ads.
- AI talking avatar video cost per minute: Sume tiers
Sume avatar video is $0.184, $0.245 or $0.55 per second by tier: $11.04, $14.70 or $33.00 a minute. 100 thirty-second clips cost $552 to $1,650.
Written by Sume