This paper presents floating point multiplier capable of supporting wide range of application domains like scientific computing and multimedia applications and also describes an implementation of a floating point multiplier that supports the IEEE 754-2008 binary interchange format with methodology for estimating the power and speed has been developed. This Pipelined vectorized floating point multiplier supporting FP16, FP32, FP64 input data and reduces the area, power, latency and increases throughput. Precision can be implemented by taking the 128 bit input operands.The floating point units consumeless power and small part of total area. Graphic Processor Units (GPUS) are specially tuned for performing a set of operations on large sets of data. This paper also presents the design of a Double precision floating point multiplication algorithm with vector support. The single precision floating point multiplier is having a path delay of 72ns and also having the operating frequency of 13.58MHz.Finally this implementation is done in Verilog HDL using Xilinx