Custom containers - setting args

Inference Endpoints on the Hub
1

I’m trying to use a custom container for inference but I cannot see anywhere how to set the args for the query that starts the container’s serving? Anyone know what i am missing?

e.g. i need to set things like this:

    f"--model={model_id}",
    f"--tensor-parallel-size={accelerator_count}",
    "--swap-space=16",
    f"--dtype={dtype}",
    "--gpu-memory-utilization=0.9",