We’re excited to announce that Vidu Q2, the latest video generation model from Vidu, is now available in ComfyUI! With breakthrough character consistency, 3x faster speed, and support for up to 7 reference subjects in a single workflow.
Model capabilities
Vidu Q2 focuses on controllability and stability—so your outputs are easier to steer and more consistent across iterations.
Core capabilities
Stronger multimodal understanding: Better at multi-step instructions, causal intent, and tone/emotion, so actions connect more naturally.
Higher-fidelity dynamic rendering: Smoother large motions with more believable physical feedback; fewer breakdowns and sudden shape drift.
Expressions + micro-movements: Finer facial expressions, eye movement, and subtle gestures for more expressive characters.
Camera language control: Push/pull/orbit/follow/close-up strategies with more
coherent transitions.Multi-character + scene coordination: More stable character interaction and complex scene composition when shots or scenes change.
Multi-image reference (up to 7 references)
Identity consistency: Face, hair, outfit, and small details
Scene consistency: Layout, lighting logic, and overall style
Narrative consistency: Stability when multiple subjects appear together
Examples
Reference to Video
Text to Video
Image to Video
Getting Started
Update ComfyUI: Ensure you’re running the latest version of ComfyUI or visit Comfy Cloud!
Access the workflows: Click the Templates in the sidebar. → Template library → search “Vidu 1Q2” workflows
As always, enjoy creating!
You should ALWAYS put when it's an API plugin.... some of us feel TRICKED of CLICK-BAITED when there is no mention and it's ONLY PAY to PLAY... would rather NOT get the newsletter if it's hidden API plugs. #Bullshit!
I do like the fact that you're rolling out new features on a weekly, if not daily basis.
But I would rather see improvements being made to the ComfyUI platform itself. Especially the PC desktop version.