Scrybl runtime: DirectML + ONNX Runtime (Windows x64)
The DirectML universal GPU runtime bundle Scrybl ships with โ works on every modern Windows GPU (NVIDIA, AMD, Intel) as a vendor-neutral fallback. Installed installer-bundled today; in a future release Scrybl Slim installs will pull this bundle from this repo on first launch instead of bundling the bytes inside the installer.
Files
- โ Microsoft DirectML runtime, Microsoft.AI.DirectML NuGet redistributable, MIT-equivalent terms.
- โ ONNX Runtime 1.24.2, MIT licensed.
- โ ONNX Runtime shared provider, MIT licensed.
Bytes are byte-identical to the upstream Microsoft / ONNX Runtime distributions. Attribution is unchanged.
Provenance
- Microsoft DirectML: github.com/microsoft/DirectML, distributed via NuGet.
- ONNX Runtime: github.com/microsoft/onnxruntime, tagged release 1.24.2.
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐ Ask for provider support