In order to realize fast and accurate Monte Carlo simulation of neutron and photon transport problems, new algorithms have been studied, and two fully vectorized Monte Carlo codes MVP and GMVP have been developed for FACOM VP series vector supercomputers. The former and latter codes use the continuous energy model and the multi-group model, respectively. Compared with conventional scalar codes, these codes achieve higher computation speed by a factor of 10 or more on FACOM VP-2600. Both codes have functions for production use including accurate physics model, geometry description capability and variance reduction technique. In addition to the execution on the vector supercomputers, these codes can be used on vector-parallel supercomputers with parallel calculation capability and on workstations with an OS of UNIX. For the continuous energy method, many nuclear data libraries have been prepared based on mainly JENDL 3. This report describes the physics model, geometry description method used in the codes and how to use them. (author).