This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(134.4, 44%, 23%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EG%3C/text%3E%3C/svg%3E)
Hello,
I have 2 pre-trained models I would like to deploy for inference using only one Azure ML managed online endpoint. One model would generate an input for the other model. The examples available on the learning platform and on GitHub all reference only one model per endpoint.
Is it possible to load 2 models under the same online managed endpoint? Do you have any reference guide on how to do it?
Thank you
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(214.4, 74%, 30%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EAM%3C/text%3E%3C/svg%3E)
Giuseppe Greetings & Welcome to Microsoft Q&A forum!
Did you check these already? Create a multimodel deployment and Create a multimodel deployment using a custom container
Also, see Azure Machine Learning SDK (v2) examples for more details,
Thank you for the links @AshokPeddakotla-MSFT, these are very helpful. I will go through the resources and try my deployment. Feel free to delete the question since I overlooked the repository
Giuseppe Glad to heat it was helpful.
If the response helped, please do click Accept Answer and Yes for was this answer helpful.
Doing so would help other community members with similar issue identify the solution. I highly appreciate your contribution to the community.
Hi @[AshokPeddakotla-MSFT], the multimodel deployment guide shows how to deploy using 2 models which are stored in the local device and then registered as a single model on azure.
Is it possible to deploy the endpoint using multi models that are already registered on azure without having to packaged them as a single "registered model"?
What I mean is that referencing two (or more) models during deployment as assets already available on azure (registered models on workspace).
Please let me know if not clear.