I noticed poetry can be a problem in my run.
Not specifically due to the cache, but due to the installation of much more packages than the runtime might need.

When using regular pip, it will use the requirements list determined by ClearML to install necessary packages, which usually already excludes all dev-tools and similar.

I am not sure if poetry uses the cache properly, but I can't verify either atm.

When developing I use the poetry environment, but in the queues I let clearML handle the installation via pip.

There is no need to use poetry if the task is a one-time thing

@<1691620877822595072:profile|FlutteringMouse14> , what version of the agent are you using?

@<1694157594333024256:profile|DisturbedParrot38> thanks a lot! I didn't know it. On my project, clearml can't recognize the requirements from the code, I have to pass them explicitly, either with poetry or with a requirements.txt. Does that work for you?

@<1523701070390366208:profile|CostlyOstrich36> It's 1.8.0

@<1691620877822595072:profile|FlutteringMouse14> , can you try upgrading to the latest available? 1.8.1rc2