Voxel drawing perofrmance for 128x128x128 grids

I am looking for an easy way to display Pix3D dataset which consists of images and voxels models as labels. I have a problem with displaying 128x128x128 Numpy arrays as models. It works but is super slow, it takes 1-2 minutes to draw the model and then the window is unresponsive.

Is this limitation matplotlib or am I doing something wrong? This is my code (input is tensor from Tensorflow 2):

    def show_voxel_model(self, model):
    fig = plt.figure(figsize=(10, 10))
    mmm = model.numpy()
    ax = fig.gca(projection='3d')
    ax.voxels(mmm, facecolors='red', edgecolor='k')