e5-mistral-7b-instruct embedding and inference in the same VLLM instance, is it possible?

[micuentadecasa]

Is it possible to run an instance of e5-mistral-7b-instruct with VLLM and use it at the same time for embedding and inference?
I don't' want to instantiate two instances that will exhaust my GPU, so it will be great if the same instance can be used for both tasks.