Hi! I’m trying to use the Wan2.2 FUN-Control model with the official ComfyUI workflow (Wan2.2 FUN-Control HIGH/LOW checkpoints).
However, after I add additional control signals (e.g., depth / canny sequences), the SVI 2.0 Pro LoRA seems to have no effect at all. The video can still be generated and clips can be stitched, but I don’t see the SVI Pro improvements (better long-term consistency / less color drift).
Question: Is SVI 2.0 Pro officially compatible with the Wan2.2 FUN-Control base model?
If not, is there a recommended way to combine FUN-Control conditioning with SVI Pro (or a specific workflow / settings that are required)?
I tried multiple times, but whenever I use FUN-Control + extra controls, SVI Pro LoRA appears to be completely inactive.
Any guidance would be greatly appreciated!