Feature Request: @-Reference Tagging for Images

ChatGPT’s image editing workflow would benefit from a Higgsfield-style @ reference system that allows users to explicitly target attached images in prompts.

Example:

  • @ref1 = face source

  • @ref2 = clothing source

  • @ref3 = environment source

  • @ref4 = lighting source

This would eliminate ambiguity in multi-image editing tasks such as face swaps, character turnarounds, style sheets, vehicle redesigns, prop design, and continuity work. The current “reference 1/reference 2” approach becomes cumbersome when many images are attached. Explicit @ tagging would improve precision, consistency, and editing reliability.

Hey @WinstonWilliams! I can see how this would help a lot.

An explicit reference-tagging system would be really helpful for more detailed image workflows, especially when users are combining faces, outfits, environments, and lighting from different sources. I can’t share a timeline right now, but I’ll pass this feedback along internally.

- Sunny