Vectorize matrix-vector multiplication

I would like to multiply 3 vectors of size 2 by a 2x2 matrix multiplication without using a loop.
With a loop it would look like this:

b=torch.tensor([[3.,4.], [5.,6], [7.,8.]])
res = []
for i in range(b.shape[0]):
    res.append(torch.matmul(A, b[i]))

but I would like is something lile torch.matmul(A, b).

I tried torch.matmul(A.repeat((2,1,1)), b) but obviously it did not work.

You can do this.