GPU-Accelerated Linear Algebra Library supports Fermi-based NVIDIA GPUs.

EM Photonics    Newark, DE 19711
Jul 16, 2010 CULA v2.0 features LAPACK interface comprised of over 150 mathematical routines, including system solvers, least squares solvers, orthogonal factorizations, eigenvalue routines, and singular value decompositions. With different interfaces available to integrate into existing code, programmers can call program from C/C++, FORTRAN, MATLAB, or Python codes. Software is available for systems equipped with GPUs based on NVIDIA CUDA architecture and with 32- and 64-bit Linux, Windows, and OS X.