I use ALGOR for many years.

It does support a long list of (nonlinear) material models, including several for rubber and rubber-like materials.

You can define for Mooney-Rivlin 2, 5 and 9 order curves (but you  usually will get data for 2-order only, C1 and C2 constants). There is even a curve fitter in the software for your own measurements.

For Mooney-Rivlin the solution converges fast and many element types are supported. It's fun.

the problem with metal to rubber bond lines using fea is the solution for the rubber and the solution for the metal material are smeared together at the bond line(nodes) giving incorrect results. The bondline strain of the rubber at the bond line is the result of interest. ADINA software by Bathe at MIT who did his PHD at UC Berkely on rubber and MARC now owned by MSC software are the best for analyzing rubber to metal. These software tools separate the solutions from the two materials. Algor does not measure the stretch of the rubber it only calculates the Von Mises stress and strain which has nothing to do with the rubber, it is used to measure behavior of metal. Also Algor does not have an element formulated for analyzing rubber. The correct element is called a mixed formulated element which incorporates a hydrostatic degree of freedom at the corners
whcih used to calculate the change in shape. Ruber does not change in volume but changes shape. Rubber has a bulk modulus close to and approaching .5.  A material with a bulk modulus of .5 has no change in volume when loaded, but has a shape change.
feapro@aol.com