Thanks for your reply. For the fitting issues, this is what I got using MAT_077_HYPERELASTIC with positive stress strain values (which is not correct because the experiment result is from compression tests). You can see in the second picture that the fitting went terrible when switch both stress and strain value to negative.

When I manually fitted the negative stress-negative strain and input C10 C20 C30, LS-DYNA shows this error:

*** Error 40009 (SOL+9)
     The time step size is zero.

I have double checked that my timestep setup are correct. I don't know how to resolve this issue.

Is it normal that using an odgen/hyperelastic/rivlin model will make the simulation to be super slow?

Another question is that if I use MAT_024_PIECEWISE_LINEAR_PLASTICITY, where the input stree-strain curve is effective stress versus effective plastic strain, which are supposed to be positive, how can I import my compression test result to it?