Symphony Seedance 2.5 is invite-only: the API route today
TikTok Symphony with Seedance 2.5 is limited to select paid advertisers in select markets. The Sume API runs seedance-2.5 for 4 to 30 seconds today.

TikTok Symphony with Dreamina Seedance 2.5, dated August 3, 2026, is offered only to select paid advertisers in select markets. If you are not one of them, the Sume API accepts seedance-2.5 for 4 to 30 seconds at 480p, 720p or 1080p through POST /v1/videos.
What TikTok announced
The figures below come from the TikTok for Business post, read on 2026-10-03. Check availability for your account and market there.
| Item | Before | Now |
|---|---|---|
| Generation length | 15 s | Up to 30 s |
| Multimodal references | 9 | Up to 50 |
| Disclosure | - | AI labels plus invisible watermark |
| Access | - | Select paid advertisers, select markets |
What Sume documents for the same model family
Limits on Sume are per model. seedance-2.5 accepts 4 to 30 seconds at 480p, 720p and 1080p; seedance-2 tops out at 15 seconds. Audio and video references are honored by the Seedance 2.x models. The reference counts above are TikTok's and Sume does not claim the same: the legacy Video 1.0 docs list 1 to 9 image, 1 to 3 video and 1 to 3 audio references, with audio requiring at least one image or video reference. Read supported_input_references from GET /v1/videos/models for a model's real inputs.
Sume does not describe a watermark or label on its outputs in these docs. If your channel requires disclosure, add it yourself.
Submit a clip
Use Idempotency-Key so a retry returns the original job. Poll the returned polling_url until the status is completed, then download from unsigned_urls.
curl -X POST https://api.sume.com/v1/videos \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: seedance-25-001" \
-d '{
"model": "seedance-2.5",
"prompt": "A vertical UGC-style product clip on a desk, natural light",
"resolution": "720p",
"duration": 15
}'Cost and queueing
The workspace USD balance is reserved on submit at provider list price times 1.25, and usage.cost on the poll response is the billable amount. Extra jobs beyond your processing concurrency are accepted as queued while queue capacity remains.
Sources
Related posts
More in Comparisons
- sync-3 at $0.133 a second vs Sume quality tiers
Sync lists sync-3 lipsync at $0.133 a second and native 4K. Sume avatar video has standard, plus and max tiers, and face swap is a Beta for short sources.
- 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.
- 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.
- TTS latency numbers side by side: Eleven v4 Turbo, Voxtral, MAI Flash
Three vendors, three latency figures, three different things measured. A table of what each page says, and why a Sume TTS job is a different question.
Written by Sume