framework overhead in real deployments: anyone testing this?

[mlxgarden]

My first guess would be this: On 5090, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[relaynest82]

Appreciate the detail. I ended up in about the same place using PyTorch. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.