Seedance 2.5 🧠 Model Paid
ByteDance's video model: 30-second clips with audio, many references and local edits
- Maker
- ByteDance Seed
- Latest release
- 2026-07
- Weights
- Closed
- Licence
- –
- Apps offering it
- 26
Where you can use Seedance 2.5
Apps in this directory that let you generate with Seedance 2.5, from their own documentation. You can also call it directly through the maker's API.
Good for
About Seedance 2.5
What it is
Seedance is the video model family from ByteDance's Seed team. Seedance 2.0 arrived in February 2026. Seedance 2.5, released in July 2026, doubles the single-generation length to 30 seconds and generates audio and video together. ByteDance ships it in Jimeng AI, Doubao and Dreamina, and the CapCut family of apps uses Seedance too.
Key features
- Up to 30 seconds in one pass, with multi-round extension. A beta long-video mode is reported at up to 3 minutes
- Up to 30 images, 10 videos and 10 audio clips as references in one request
- Timestamp-level editing: fix one detail without regenerating the whole clip
- Green-screen background replacement and camera-angle changes
- Seedance 2.0 lip-syncs in 8+ languages (per Freepik)
Where you can use it
ByteDance's apps: Jimeng AI, Doubao Pro and Dreamina, plus CapCut (Dreamina Seedance 2.0). Seedance 2.0 or 2.5 is documented in Higgsfield, Krea, Freepik, Leonardo.Ai, Runway (Seedance 2.0 Pro and Fast), Luma's app, invideo AI and ComfyUI partner nodes. ByteDance says API access to 2.5 is coming via BytePlus ModelArk.
Pricing and rights
Seedance is sold through app credits, not a standalone subscription. Prices depend on the app (CapCut, Dreamina, Higgsfield and others), so price_from is left empty. ByteDance's launch post does not cover resolution tiers, watermarking or rules on real-person likeness. Check the terms of the app you use, especially for ads.
Who it is for
Short-drama and ad creators who need longer single takes, consistent characters and many references. CapCut users who want generation inside their editor.
Verdict
Seedance 2.5 leads on clip length and reference control, and it is spreading fast through creator apps. Official specs and API pricing for 2.5 were still thin at launch. Access depends on which ByteDance or partner app you use.
Pros
- 30-second single generations
- Very rich multimodal references
- Local edits without full regeneration
- Available in CapCut and many multi-model apps
Cons
- 2.5 API not generally available at launch
- Sparse official specs (resolution, watermark policy)
- Pricing scattered across apps
Similar AI models
All video models →Veo 3.1 🧠 ModelPaid
Google DeepMind's text- and image-to-video model with native sound, up to 4K
Kling 3.0 🧠 ModelFreemium
Kuaishou's video model with native multilingual audio, lip sync and 15-second clips
MiniMax H3 (Hailuo 3.0) 🧠 ModelFreemium
MiniMax's open-weight video model: native 2K, stereo sound, 15-second clips
Runway Gen-4.5 🧠 ModelFreemium
Runway's own video model, now with native audio and one-minute multi-shot scenes
Wan 3.0 🧠 ModelOpen source
Alibaba's video family: open-weight Wan 2.x and the 30-second Wan 3.0 API
LTX-2.5 🧠 ModelOpen source
Lightricks' open-weight video model with synchronised audio and 4K output
More from ByteDance Seed
Seedream 5.0 Pro 🧠 ModelFreemium
ByteDance's Seedream image models, used in CapCut, Dreamina and many creative apps