Best practice of sparsify and quantization

[lnexenl]

Hello I am going to deploy a model on Thor-U. I have already quantized the model, but I am confused what should I do to combine quantization and sparsify together? Is there any best practice for doing these together? Thank you 🥹