Milestones

List view

• Make generation independent from requested region

  Right now, current implementation output depends on region location and size. To make diffusion truly infinite, we need to make requested regions independent of where windows are generated

  No due date

  •0/3 issues closed
  0% complete3 open0 closed

• Get a barebones end to end functional workflow

  ### Ask Create a static image from noise and display to the frontend. ### Milestones - Generate a static image of fixed size - Split image into subregions - Run one processing step over each subregion (e.g. blur) - Merge resulting post-processed static into one image - Set up API endpoint to request images from backend - Send and display static image to frontend ### End Requirements We open the frontend and see a different static image on every page load, randomly generated and processed by the backend.

  Overdue by 13 day(s)

  •

  Due by August 15, 2026

  •19/41 issues closed
  46% complete22 open19 closed