Synthesia Style Avatars from a prompt: the Sume avatar input types

Synthesia's Avatar Builder creates Style Avatars, which its launch post describes as stylized, branded avatars, from a prompt and a preset, with reference images listed as coming soon. Sume has three avatar input types on POST /v1/avatar-1.0/generate: a Prompt, a Profile and an Image. It documents no preset input.

Synthesia facts are from its launch post. Sume facts are from Create new avatar, read 2026-10-01.

What is a Style Avatar?

Per the post, Style Avatars are stylized cartoon and character avatars powered by Express-3, aimed at brand mascots and at topics where a human presenter may not land. You name one, pair it with a voice and use it in Assistant.

How do Sume's three inputs compare?

Each request is a job. Poll it, then use the returned avatar handle or resource id to generate avatar videos. The handle may include a leading @; Sume stores it without one.

What must an image URL satisfy?

image_url must be a fetchable public HTTPS image URL. Localhost, private-network URLs, non-HTTPS URLs and non-image responses are rejected before generation is submitted. See the credits-per-option comparison for how an avatar job is billed.

Can I get a cartoon or mascot look?

Describe it in the Prompt input and inspect the result. The docs do not list a style catalog, so treat the look as something you iterate on. Send an Idempotency-Key so a retried create does not make a second avatar.