Hi everyone,
I’m wondering if there’s a workflow in ComfyUI that works similarly to the Moodboards feature in Midjourney or Krea.
My goal is to:
- Collect a set of reference images (around 5–20 images).
- Use those images as a visual moodboard to define the overall style, lighting, colors, composition, and materials.
- Generate new images that follow the same overall style, rather than copying any single reference image.
I know I can provide multiple image inputs to models like Nano Banana, but I’m looking for something closer to how Moodboards work in Midjourney or Krea.
Is there an existing workflow, custom node, or recommended approach for this in ComfyUI?
If not, how would you build a similar pipeline?
My goal is to create consistent keyframes for AI video production, where I can simply swap the moodboard images and generate a new visual direction.
Thanks in advance!