bluzen/prismstream-xl
Text-to-Image
Flow-matching text-to-image model with region conditioning. The reference checkpoint for the PrismStream stack.
Controllable generation
The PrismStream line: the checkpoint, the paired line-art data behind its conditioning, and the demo where you can paint a region and argue with it.
All collectionsUpdated 20 August 2026
Models(2)
bluzen/tagflow-350m
Text Generation
Tag-to-prompt expansion with constrained decoding, so every tag you supply survives into the output.
Datasets(1)
bluzen/lineart-pairs
Image-to-Image
Aligned line-art and colour pairs for conditioning experiments, extracted with a published pipeline.
Spaces(2)
prismstream-demo
bluzen/prismstream-demo
Text-to-Image
Text-to-image with region conditioning. Paint a mask, describe what belongs inside it, watch it argue with you.
prompt-expander
bluzen/prompt-expander
Text Generation
Type a handful of tags, get a full prompt back with every tag you supplied still in it.