Seedance 2.5
#16 in Text-to-VideoUnknown · v2.5 · since 31. Juli 2026 (Vorschau 23. Juni 2026) · 2× · last seen Oct 09, 2026
36
Momentum
Seedance 2.5 is a text-to-video audio-video joint generation model from ByteDance Seed. According to the vendor, it generates native videos up to 30 seconds in a single pass and accepts up to 50 references (30 images, 10 videos, 10 audio files). The model is proprietary (not open source); access is via Dreamina/Jimeng, Doubao Pro and third-party platforms and API providers.
Momentum trend
11.07.09.10.
Features
| Fine-Tuning | No fine-tuning; instead reference control (up to 50 references) and region-level editing |
| License | Proprietary, not open source; use governed by platform terms (Seedance pricing FAQ: commercial use in all plans) |
| Max Resolution | 1080p (platform-dependent; default output 720p/480p, 4K not confirmed by sources) |
| Max Video Length | 30 seconds per generation (4 to 30 s) |
| Platform | Jimeng AI, Doubao Pro (official); API via BytePlus ModelArk |
| Price | Dreamina campaign: from $0.035/s at 720p (Basic, first month, 23 Sep–9 Oct 2026); BytePlus ModelArk: $0.231/s at 720p, $0.103/s at 480p (no video input) |
| Release Date | July 31, 2026 |