Experimental Blackwell build for triton and python 3.12, to work on Blackwell 2.0 (RTX 5000 series) on Windows for nightly builds with cuda 128.
Also works with Ampere and Ada.
Mostly made it to work on reforge for windows, but it should work on every place as well (or maybe not)
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.