[Question]: Is OpenVINO EP from the Windows ML catalog supported with the new Runtime API and ORT 1.30.0 (or maybe 1.31.0)?
Nobody has claimed this yet.
Assessment
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Newbie friendliness
- 68/100
- Issue type
- Documentation
- Clarity
- Clearly specified
- Activity status
- Active
- Domain
- documentation, machine-learning
Research direction
Start with the linked Windows ML Runtime API overview, supported execution providers page, and EP release history; compare those with the referenced OpenVINO NuGet package details. Clarify whether the catalog OpenVINO EP works with Runtime API and ORT 1.30.0, how to register it, and what runtime versions the cited packages bundle. Done when the compatibility answers are documented, ideally with the requested C++ example and version matrix.
Written by the indexing model from the issue text.
Description
What's your question?
I'm evaluating the new Windows ML Runtime API and would like to confirm whether the OpenVINO Execution Provider distributed through the Windows ML ExecutionProviderCatalog is supported with Microsoft.Windows.AI.MachineLearning 2.7.2021-experimental, which ships ONNX Runtime 1.30.0.
I understand that the new Runtime API uses ONNX Runtime as its backend for .onnx and .ort models, and that Windows ML can dynamically acquire and register OpenVINO EP packages. However, I could not find an explicit compatibility statement or end-to-end example showing OpenVINO EP selected through the new Runtime API.
Could you please clarify:
Is the Windows ML-managed OpenVINOExecutionProvider supported with the new Runtime API and ORT 1.30.0?
If supported, what is the recommended way to register and select the provider through the new Runtime API?
The Windows ML EP release history lists package 1.8.84.0 with “OpenVINO 1.6.1” and package 1.8.80.0 with “OpenVINO 1.4.1”. Are these values the versions of the NuGet package Intel.ML.OnnxRuntime.EP.OpenVINO, rather than the underlying OpenVINO Runtime? If so, which OpenVINO Runtime versions are bundled in those packages?
Has the OpenVINO EP package 1.8.95.0 been validated with ORT 1.30.0? If so, which OpenVINO Runtime version does it contain?
A minimal C++ example and a compatibility matrix covering Windows ML, ORT, the Windows ML EP package, and the bundled OpenVINO Runtime would be very helpful.
This is a compatibility and documentation question; I have not yet established that this combination fails at runtime.
References:
https://learn.microsoft.com/en-us/windows/ai/new-windows-ml/runtime/overview
https://learn.microsoft.com/en-us/windows/ai/new-windows-ml/supported-execution-providers
https://github.com/microsoft/WindowsML/wiki/Windows-ML-Execution-Provider-Releases
https://www.nuget.org/packages/Intel.ML.OnnxRuntime.EP.OpenVINO
Related sample (if any)
No response
Context
No response
- Dominant language
- PowerShell
- Stars
- 41
- Forks
- 6
- Avg merge
- 6m
- Merged PRs (30d)
- 1
Getting set up
- No Dockerfile or Docker Compose file
- Has a pull request template
- No contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from microsoft/WindowsML
-
Difficulty 4/5 3-5 days Newbie friendliness 48/100
-
question
Difficulty 4/5 3-5 days Newbie friendliness 35/100
-
bug
Difficulty 5/5 Over a week Newbie friendliness 35/100
-
Known issue: Microsoft.Windows.AI.MachineLearning nuget fails with .NET PublishSingleFileMay be free again @zadesai claimed this 103 days ago, and no pull request is open. Open
All issues in microsoft/WindowsML
Similar issues
-
README for cliPossibly taken @Janvi-kapoor claimed this today. OpenOSCI'26
Difficulty 2/5 1-3 hours Newbie friendliness 70/100
GauravKarakoti/SecureFlow#1308 · 1 comment ·
Maintainers usually reply within 1 day
-
documentation
Difficulty 2/5 1-3 hours Newbie friendliness 62/100
vitest-dev/vitest#11577 ·
Maintainers usually reply within 1 day
-
sync-en
Difficulty 2/5 1-3 hours Newbie friendliness 62/100
Maintainers usually reply within 1 day
-
sync-en
Difficulty 2/5 1-3 hours Newbie friendliness 66/100
Maintainers usually reply within 4 days
-
[oblt-aw][autodoc] Update user onboarding guide for mandatory trigger CODEOWNERS stepPossibly taken @github-actions claimed this today. Open
Difficulty 2/5 1-3 hours Newbie friendliness 66/100
Maintainers usually reply within 1 day