A video pipeline separating planning, Seedance 2.5 generation, and conversational CapCut editing

HeyZoyaKhan · x · 2026-09-09

A creator demonstrates a workflow that separates generation from editing:

The core idea: lock down visuals in a controllable 3D layer, hand generation to the model, then iterate in natural language — three decoupled, controllable stages.

Related event: New AI Video Pipeline Separates Planning, Generation and Editing(4 posts)→

Original post →

More from Multimodal

Multimodal channel →