I am using some deep learning model inside a task in the flow. The flow is configured with docker agent. It is downloading the model from web everytime I run the flow. How to load the model from local system where the agent is running?

How would you do it in Python? Once you figure out how you would solve it in Python, we cal help you to build a Prefect flow out of it - feel free to share your code Generally speaking, Prefect flows are just Python functions decorated with task and flow decorator - this page may help you get started with flows, but if you have some more specific questions, please ask If you are asking how to load a file from a host machine from a Docker container running the flow, you would need to mount the volume - I assume you are asking for Prefect 1.0 - in that case, you could bind mount the volume when starting the agent:

prefect agent docker start --label AGENT_LABEL --volume /local/path:/container/path

imagePullPolicy if deoloying in k8s