Many recent research efforts have been devoted to the use of commodity hardware for solving computationallyintensive scientific problems. Among such problems, hyperspectral imaging has created new processing challenges in the remote sensing community. Hyperspectral sensors are now capable of collecting hundreds of images, corresponding to different wavelength channels, for the same area on the surface of the Earth. For instance, NASA is continuously gathering high-dimensional image data with hyperspectral sensors such as Jet Propulsion Laboratory?s Airborne Visible-Infrared Imaging Spectrometer (AVIRIS). The increasing programmability and parallelism of commodity graphics processing units (GPUs) makes them strong candidates for addressing some of these challenges. In this paper, we describe a GPU-based framework for implementation of hyperspectral image processing algorithms which takes advantage of multiple levels of parallelism found in modern GPUs. This framework is inexpensive, uses readily available PC graphics hardware boards, and provides the desired performance at the quality required. Experimental results are presented and discussed in the context of a realistic application, based on hyperspectral data collected by NASA?s AVIRIS system.