TikTok product avatars skip shoes, hats, sunglasses and bracelets
TikTok's Symphony help page lists products its avatars cannot show. What to try for those SKUs, and what Sume's avatar product_image does and does not claim.
TikTok's help page for avatar videos in Symphony Creative Studio says clothing accessories such as shoes, hats, sunglasses and bracelets ("etc.") are not supported for product avatars, so a holiday shoe or accessory SKU needs another route. Sume's avatar talking video takes an optional product_image, and its Formats start from a product still, but neither carries a guarantee for those products. Test one clip per product type before you commit a batch.
What does TikTok say?
Read from TikTok's own help page on 2026-10-01:
| Item | On the page |
|---|---|
| Unsupported for product avatars | Clothing accessories such as shoes, hats, sunglasses, bracelets, etc. |
| Labelling | An AI-generated label is added to all exported videos |
| Requirements | A TikTok for Business account and acceptance of the Creative GenAI Terms |
What can I try on Sume for those products?
Two routes are documented. The avatar talking video, POST /v1/avatar-1.0/talking-video, takes an avatar_handle, exactly one of script or video_inputs, and an optional product_image. The planned duration is 4 to 60 seconds. A catalog Format such as sume-close-camera-ugc builds a clip from a product still. Neither page lists shoes or hats as supported or unsupported, so a first test is the only evidence.
curl -X POST https://api.sume.com/v1/avatar-1.0/talking-video \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: avatar-hat-001" \
-d '{
"avatar_handle": "my_creator",
"script": "This hat is my gift pick this year. It fits every head.",
"product_image": "https://media.sume.com/artifacts/artf_demo/hat.png",
"aspect_ratio": "9:16"
}'How do I check the result?
Watch for the product shape in the frame, especially across a cut. Hats, glasses and bracelets are small and sit close to a face or hand, which is where generated video drifts. If the product changes between frames, shorten the clip or try a Format that holds a single still reference instead of a talking presenter.
What is a fair test plan?
Take one SKU from each excluded type and run the same short brief three ways: an avatar clip with product_image, a catalog Format clip from the same still, and a plain still with a timeline hold. Score them on one question, whether a stranger can tell what the product is. That tells you which route to use for the rest of the type, and it costs three small jobs.
What are the limits?
- TikTok's list is about TikTok's product avatars, not a rule for any other tool.
- The handle in the example is a placeholder; use an avatar your workspace owns or that the API lists.
- The AI-generated label is TikTok's behaviour on its own exports; check how a clip you upload from elsewhere is labelled, and follow the platform's disclosure rules.
- Do not use a real person's likeness without permission.
Sources
Related posts
More in Comparisons
- Together AI video API alternative: Sume /v1/videos compared
Together AI's video API is create-then-retrieve with five statuses. How that maps to Sume /v1/videos, and what each side does better.
- Veo 3.1 1080p 8-second clip cost vs Sume per-second rows
What an 8-second 1080p clip costs on Veo 3.1 Standard, Fast and Lite list prices versus five 1080p rows in the Sume video catalog, dated 2026-10-01.
- Veo 3.1 4K price per second vs Gemini Omni Flash 4K
Veo 3.1 4K costs $0.60 per second on Standard and $0.30 on Fast. Gemini Omni Flash 1.1 on Sume lists 4K at $0.375 per second. Dated 2026-10-01.
- Veo in Google Ads: 3 images, 10 seconds, and the Sume equivalent
Veo in Google Ads takes up to three images and returns videos up to 10 s. Here is the same job on Sume's video router, with its limits, for use outside Google.
Written by Sume