midjourney

Qwen Image Edit 2511: Better Multi-Person Editing

Qwen Image Edit 2511 introduces enhanced multi-person editing capabilities, allowing users to modify multiple individuals within a single image with improved

Someone checked out Qwen’s Image Edit 2511 model and found some pretty solid upgrades for anyone doing image editing work.

The model is at https://huggingface.co/Qwen/Qwen-Image-Edit-2511 and handles stuff that usually breaks other models - like keeping multiple people looking consistent in group shots, or maintaining someone’s actual face during edits instead of drifting into a different person halfway through.

What improved:

  • Multi-person scenes don’t turn into a mess anymore
  • Popular LoRAs are baked in (no fine-tuning needed)
  • Better at geometric stuff like construction lines for design work
  • Identity consistency actually works now instead of morphing faces

Turns out this version is pretty useful for product design and engineering workflows where precision matters. The geometric reasoning upgrade means it understands structural edits better than previous versions.

Free to test on Hugging Face if anyone wants to compare it against other image editing models.