HeyGen look packs vs Sume: a new avatar per look
HeyGen's POST /v3/avatars/looks adds looks to an avatar. Sume has no look-pack endpoint: make one avatar per look and steer the scene with a prompt.
Sume has no look-pack endpoint. To get several looks, create a separate avatar for each one from a prompt, a profile or a reference image, each with its own handle, and use a scene prompt for scene direction in each video.
HeyGen details are from its API changelog; Sume details from Create new avatar and Avatar videos, read 2026-09-30.
What is a HeyGen look pack?
Per the changelog, POST /v3/avatars/looks generates new looks for an avatar you already own. You pass one look you own as reference_look_id; results join that look's character. Their ids come back immediately, but video generation must wait until the looks finish processing. A Look Pack is a curated set built around one theme, and the pack decides how many looks it makes.
How does Sume model it?
| Need | Sume approach |
|---|---|
| A new look | A new avatar via POST /v1/avatar-1.0/generate with its own avatar_handle |
| Inputs | Prompt, Profile, or Image (a public HTTPS image_url) |
| Result | Each request is a job; poll it, then use the returned handle |
| Setting per video | scene: { "type": "prompt", "prompt": "..." } |
| Several avatars in one video | Not supported: one resolved avatar per final video |
How do I get a set of looks?
Submit one create request per look, each with a distinct handle such as presenter_casual and presenter_formal, and an Idempotency-Key. Pass a reference image per look if you have one. Then pick the handle per video. This produces separate avatars, not one character with variants, and the docs do not promise identity matching between them.
When is the scene prompt enough?
If you only need a different room or direction for the same avatar, leave the avatar alone and change the scene prompt. For a retouch or outfit change on a photo you already have, see HeyGen edit look versus Sume photo input.
Sources
Related posts
More in Use cases
- Listing video with an AI avatar: HeyGen for Real Estate, and Sume
HeyGen launched a real estate offer for listing videos. With Sume you can build one from an avatar, a property photo as the scene and a 4-60 second script.
- Hinglish text to speech API: romanized Hindi with language hi
For romanized Hindi or Hinglish on Sume TTS, send language `hi` even though the text is Latin script, and pin a Sonic model id on the Router.
- Holiday ad creative by phase: four videos, one bulk run
Meta splits the holidays into discovery, deals, gifting and fresh start. Queue one video per phase as a single Sume bulk run with a spend cap on each.
- India's synthetic-media label rule: does routine editing count?
India's amended IT rules exclude routine editing, formatting and compression from synthetic media, and require a prominent label on the rest. What Sume offers.
Written by Sume