EditVid: unified training-free video editing for instruction- and subject-guided edits

PLAN-Lab · hf · 2026-09-07

PLAN-Lab released EditVid on Hugging Face, a unified training-free video editing framework. It combines sparse causal memory, token injection, and soft latent blending to support both instruction-guided and subject-guided edits with high fidelity.

Original post →

More from Multimodal

Multimodal channel →