Matrix rank using vmap

I am wondering if it is possible to compute the rank of a matrix of a list of tensors with different shapes using vmap.
Every attempt I make ends in failure.

Please help.

I found promising avenue using torch.nested.nested_tensor, but the following pops up:

RuntimeError: Internal error: NestedTensorImpl doesn’t support sizes. Please file an issue on GitHub - pytorch/nestedtensor: [Prototype] Tools for the concurrent manipulation of variably sized Tensors.

Unfortunately, the given repository is already archived.