Synthesia Pro has 5 personal avatars; Sume avatars are handles
Synthesia Pro lists 5 personal avatars and Enterprise unlimited. On Sume an avatar is a reusable handle made by a job, so plan your roster by brand.
Synthesia's pricing page lists 5 personal avatars on the Pro plan and unlimited personal avatars on Enterprise. Sume has no slot count in its docs: an avatar is a reusable resource created by POST /v1/avatar-1.0/generate from a prompt, a profile or a reference image, and you refer to it by a handle in later talking-video calls. For a team with several brands, that changes the question from how many slots to how to name and organise handles.
Synthesia source: pricing, read 2026-10-04. Sume sources: Create new avatar and Generate avatar video.
What the plan limit means on each side
On Synthesia the count is a plan feature, so a sixth personal avatar on Pro means looking at a different plan; the page lists unlimited personal avatars only on Enterprise. On Sume the docs describe the avatar as a job-backed resource in your workspace and do not state a cap, and this post does not invent one. Check your account limits and the pricing for avatar creation before you commit to a large roster.
| Topic | Synthesia | Sume |
|---|---|---|
| Plan statement | Pro: 5 personal avatars; Enterprise: unlimited | No slot count in the docs read |
| How an avatar is made | In the Synthesia product | Prompt, profile (props) or photo input via the API |
| How you refer to it | Avatar on the create call | avatar_handle, stored without a leading @ |
| Creation is asynchronous | Not described on the page read | Yes, job-backed; poll until completed |
Naming a roster
The handle is the stable key your code uses, so choose it like a database id. A leading @ is accepted but stored without it. Keep one handle per face and per brand voice, and avoid encoding a campaign into the name, because the same face will serve several campaigns.
brand_hostfor the main presenter.brand_host_altfor a second look.- Put the date or campaign in your own table, not in the handle.
The workflow after creation
Create the avatar once, poll the job to completed, then call the talking-video route with avatar_handle. The video docs call for a ready avatar, so store the avatar job and gate video submits on it.
curl -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: avatar-brand-host-001" \
-d '{"avatar_handle":"brand_host","input":{"type":"props","ethnicity":"Asian","sex":"female","age":28}}'Bottom line
If you want exactly five named presenters and a hosted editor, Synthesia Pro is a clear package. If you want to create avatars programmatically and refer to them by handle in a pipeline, Sume's model fits better. Either way, record who may use which face; a real person behind a photo avatar must have agreed.
Sources
Related posts
More in Comparisons
- Preview an avatar scene before the full render
Synthesia lists individual scene preview. Sume avatar video previews render first-frame stills, then start the full render from the one you approve.
- Synthesia test videos are free; Sume previews first frames instead
Synthesia's API test flag renders free watermarked drafts. Sume has no test flag; avatar video previews approve first frames before the full render.
- Tavus lists 1080p conversational video; Sume avatar clips are 720p
Tavus's pricing page lists 1080p and 24 kHz audio on all plans; Griffin-Lite renders 720p. Sume avatar videos are 720p today. What that means for your delivery.
- Is there a Tavus Griffin-Lite API? Not yet, and what to use
Tavus says Griffin is not available to customers, only to select trusted testers. Sume's avatar video API ships today with 4-60 s clips and five ratios.
Written by Sume