Seedance 2.5
#12 in Text-to-VideoArtlist · v2.5 · 3× · last seen Aug 13, 2026
47
Momentum
Seedance 2.5 is ByteDance's latest AI video model, offered through the Artlist platform as part of the AI-Toolkit. It natively generates video clips ranging from 4 to 30 seconds in length in a single pass and accepts up to 50 multimodal references (images, video clips, audio files). Audio such as dialogue, music, and effects are generated synchronously with the image rather than added post-production. At market launch, the resolution is 480p/720p at 24 fps; for higher resolutions (up to 4K), Artlist continues to recommend Seedance 2.0.
Momentum trend
15.05.13.08.
Features
| Fine-Tuning | No classic fine-tuning; instead control via up to 50 multimodal references (30 images, 10 video clips, 10 audio files) per generation |
| Max Resolution | 480p and 720p at 24 FPS at launch; for up to 4K, users are directed to Seedance 2.0 |
| Max Video Length | 30 seconds natively in a single pass (4–30 second range adjustable; Smart Duration auto-selects 4–15 seconds) |
| Platform | Available in Artlist's AI Toolkit (web platform), also via Artlist MCP integration (e.g., via Claude) |
| Price | From $41.67/mo (billed annually, $499/year) for Unlimited access at 480/720p via Artlist; alternatively usable via paid AI plans using credits (amount depends on clip length/references) |
| Release Date | Officially announced June 23, 2026 (Volcano Engine conference); live on Artlist since approx. early August 2026 |