ComfyUI Workflow Help: How to Place an Existing Character into a New Scene?

DKligerSC · reddit · 2026-08-05

A creator asked Reddit for help with an image generation workflow. They originally used Tensor.art for character portraits and Grok to place them into specific scenes (e.g., a VIP office).

Tired of paywalls, they moved to the open-source ComfyUI and successfully replicated the character generation. However, they are stuck on how to use ComfyUI (likely image-to-image) to maintain character consistency while changing camera angles, lighting, and integrating a completely new scene. They are looking for workflow guidance from the community.

Original post →

More from Apps

Apps channel →