Daoud AI Workflows

Combine three visual references for character identity, scene and expression in one controlled shot.

Included: One editable ComfyUI graph and concise setup and validation guide.
Requirements: Current ComfyUI, matching local MiniMax H3 model, Qwen3-VL encoder, video/audio VAEs, referenced custom nodes, a capable GPU, and your own visual/audio inputs.
Before buying: this is an editable ComfyUI workflow template, not rendered footage. Offline graph checks only; no clean-install render or quality guarantee yet.