Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

OpenVINO Backend 2022.3 support #5343

Closed
riZZZhik opened this issue Feb 10, 2023 · 14 comments · Fixed by #6031
Closed

OpenVINO Backend 2022.3 support #5343

riZZZhik opened this issue Feb 10, 2023 · 14 comments · Fixed by #6031
Assignees
Labels
investigating The developement team is investigating this issue

Comments

@riZZZhik
Copy link

Hello,

Currently triton-inference-server/openvino_backend is based on OpenVINO 2022.01
When will official Triton Container support 2022.3 or later OV backend?

Based on Changelog (2022.02, 2022.03) I assume that update won't cause any issues

Thanks!

@nv-kmcgill53
Copy link
Contributor

cc: @Christina-Young-NVIDIA @dzier

@riZZZhik
Copy link
Author

Any updates about this issue?

@GuanLuo
Copy link
Contributor

GuanLuo commented Mar 13, 2023

@tanmayv25 can you estimate this effort?

@tanmayv25
Copy link
Contributor

tanmayv25 commented Mar 13, 2023

The standalone openvino version in OV backend is dependent upon the version of openvino released in onnxruntime version being shipped with Triton. See here for more information: https://github.com/triton-inference-server/server/blob/main/build.py#L64-L67

The actual update is not expected to be a lot of work.

@riZZZhik
Copy link
Author

@tanmayv25 Hello,

Could you please let me know when it will be implemented

@tanmayv25
Copy link
Contributor

@pranavsharma Does the latest onnxruntime release in Triton ORT v1.14.1 support openvio 2022.3 ? If yes, then we can try upgrading the openvino version in Triton.

@pranavsharma
Copy link
Contributor

@pranavsharma Does the latest onnxruntime release in Triton ORT v1.14.1 support openvio 2022.3 ? If yes, then we can try upgrading the openvino version in Triton.

yes. https://onnxruntime.ai/docs/execution-providers/OpenVINO-ExecutionProvider.html#requirements

@siyuch-fdu
Copy link

We also need the 2022.3 version of openvino_backend, and hope that the official version can be implemented as soon as possible, thank you very much!

@nv-kmcgill53
Copy link
Contributor

CC: @nnshah1

@nikolsky
Copy link

It would be great to see this feature in the future releases.
We have to use OpenVINO Model Server instead of Triton without it.

@pranavsharma
Copy link
Contributor

The latest version of ORT uses 2023. Does that work? See https://onnxruntime.ai/docs/execution-providers/OpenVINO-ExecutionProvider.html#requirements

@riZZZhik
Copy link
Author

@pranavsharma it supports OpenVINO 2023, but uses the one that is installed in triton server
Currently it is 2022.1 as in https://github.com/triton-inference-server/openvino_backend

@nnshah1 nnshah1 added the investigating The developement team is investigating this issue label Jun 30, 2023
@nnshah1
Copy link
Contributor

nnshah1 commented Jun 30, 2023

Quick update - the team is currently investigating the upgrade.

@riZZZhik
Copy link
Author

@kthui Thanks! ❤️❤️❤️

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
investigating The developement team is investigating this issue
Development

Successfully merging a pull request may close this issue.

9 participants