SDXL Turbo Unet model for OnnxStream, see https://github.com/vitoplantamura/OnnxStream/issues/44

SDXL turbo benefits greatly from a lower step size compared to SDXL but higher step counts seem to not benefit it as much as it would SDXL

SDXLTurbo steps=1 seed=1000 SDXL steps=1 seed=1000
row01 Invalid
SDXLTurbo steps=3 seed=1000 SDXL steps=3 seed=1000
row01 row02
SDXLTurbo steps=5 seed=1000 SDXL steps=5 seed=1000
row01 row02
SDXLTurbo steps=10 seed=1000 SDXL steps=10 seed=1000
row03 row04
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.