Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Windows ML includes optional, dynamically downloaded components such as execution providers and runtime libraries that are provided by third parties. Each of these components is governed by its own license terms, separate from the license that applies to Windows ML itself. Before your app uses any of the components listed on this page, read the corresponding license terms.
Execution provider licenses
The following license terms apply to the execution providers described in Windows ML execution providers.
| Execution provider | Vendor | License terms |
|---|---|---|
| MIGraphX | AMD | Ryzen AI Licensing Information |
| NvTensorRtRtx | NVIDIA | NVIDIA SOFTWARE LICENSE AGREEMENT and License Agreement for NVIDIA Software Development Kits — EULA |
| OpenVINO | Intel | Intel OBL Distribution Commercial Use License Agreement v2025.02.12 |
| QNN | Qualcomm | To view the QNN License, download the Qualcomm® Neural Processing SDK, extract the ZIP, and open the LICENSE.pdf file. |
| VitisAI | AMD | Ryzen AI Licensing Information |
| WebGPU (experimental) | Microsoft | WebGPU EP for Windows App SDK License and ONNX Runtime License |
Llama.cpp and GGML component licenses
Windows ML also makes use of the following third-party components for Llama.cpp and GGML based inference, including CUDA acceleration.
| Component | License | License terms |
|---|---|---|
| Llama.cpp Core | MIT | llama.cpp LICENSE |
| GGML | MIT | ggml LICENSE |
| CUDA GGML backend | MIT | llama.cpp LICENSE |
| NVIDIA CUDA runtime dependencies | NVIDIA CUDA Toolkit EULA | License Agreement for NVIDIA Software Development Kits — EULA |
The NVIDIA CUDA runtime dependencies are CUDA Toolkit dependencies shipped as part of the CUDA GGML backend. This includes the dynamically loaded cuBLAS library, which your app acquires only after you set a flag indicating that you accept its license terms.