Upload an image of a subject, describe a new scene, and generate a version of that subject placed into it. Powered by OminiControl, a subject-driven diffusion model — embedded directly from its Hugging Face Space.
This runs on shared community GPU compute (Hugging Face's "Zero" hardware) — generation can take anywhere from a few seconds to a minute or two depending on current queue load. Model and app built by Yuanshi — view the original Space ↗.