Rémi Louf
Rémi Louf
Closing as duplicate of #1445
We generally use HF's internal testing models which are much smaller, see the other tests. Can you try to find if there is one for audio?
We could also attempt to perform compilation while the model weights are being downloaded + during the first forward pass that typically takes longer because of prompt processing. However, since...
Pydantic does offer [partial JSON parsing](https://docs.pydantic.dev/latest/concepts/json/#partial-json-parsing), is that something we'd want to use here?
I meant in the "output" field of the error, sorry for not being more specific.
Is there a way we could use triton's [Python client](https://github.com/triton-inference-server/client/tree/main/src/python)?
What do you mean by planned to do this another way?
I have tried to contact the package name's owner twice since I first posted this issue, to no avail.
Thank you very much for your time and help.