automatic/TODO.md

1.5 KiB

TODO

Main ToDo list can be found at GitHub projects

Current

Issues/Limitations

Future Candidates

Code TODO

  • enable ROCm for windows when available
  • resize image: enable full VAE mode for resize-latent
  • infotext: handle using regex instead
  • processing: remove duplicate mask params
  • model loader: implement model in-memory caching
  • hypertile: vae breaks when using non-standard sizes
  • force-reloading entire model as loading transformers only leads to massive memory usage
  • add other quantization types
  • lora make support quantized flux
  • control: support scripts via api
  • modernui: monkey-patch for missing tabs.select event