Seedance 2.5 Teaser: Video Length Doubled
DavidmComfort · x · 2026-07-15
Volcano Engine has teased Seedance 2.5 ahead of its debut tomorrow, highlighting four major upgrades:
- Single video duration is extended from 15 seconds to 30 seconds, making it ideal for complete dialogues, continuous actions, and one-shot storytelling.
- The reference material limit is increased to 50, allowing simultaneous inputs of characters, clothing, scenes, props, actions, and sounds to maintain a unified style for short dramas and branded content.
- Localized retouching is supported, enabling specific modifications to gestures, props, or backgrounds without regenerating the entire video.
- Enhanced camera and spatial controls, combined with action references and white/green screens, make it suitable for one-shot takes, multi-person interactions, fights, and complex blocking.
The post also mentions that Volcano Engine teamed up with English football star Michael Owen to recreate his career highlights using Seedance 2.5, showcasing the model's expanding boundaries in AI video creation.
Related event: Seedance 2.5 Debuts With a Focus on Realism in Complex Video Scenes(10 posts)→
More from Multimodal
- Getting Started with AI Video: Solving Consistency and Censorship — cynicalnewenglander · 2026-07-22
- Storyboard-first workflows are making AI dance videos and influencers more consistent — aftahi_ai · 2026-07-22
- Runpod MCP and Claude help spin up image and video generation workflows — 802high · 2026-07-22
- Midjourney prompt turns a bee into a glitching pixel explosion — michaelrabone · 2026-07-22
- A physics reward can improve video generation without creating a real physics engine — Dapper-Drawer4546 · 2026-07-22
- HeyGen adds a media-sourcing skill for coding agents with 75k images and 10k tracks — HeyGen · 2026-07-22