PerfXLab, with its core team from the Chinese Academy of Sciences, is dedicated to the research and development of computing software stack technology, gathering three major fields: HPC, AI, and RISC-V. PerfXLab is an international leading accelerated computer technology solution provider, enabling computing chips, scientific computing, and high-performance computing application software companies. In the future, perfxlab will provide the underlying and basic technologies globally, and promote the technical support of computing hardware technology and computing power basic technology services.
PerfMPL mainly includes Blas, FFT, IPP, math and other key mathematical calculation libraries and deep learning operator libraries. PerfMPL is widely adapted to arm, risc-v and x86 CPU processors. PerfMPL builds and optimizes core computing library components for NPU, GPU, DPU and other accelerators.
PerfMPL Technical White Paper
DOWNLOAD：Please register an account to get real-time updates.
PerfMPL for General Computing API Document
DOWNLOAD：Please register an account to get real-time updates.
OpenBLAS
Open Source Project: GitHub
RISC-V Computing Library (RVCL)
DOWNLOAD：Please register an account to get real-time updates.
PerfXPy is a high-performance heterogeneous python computing platform for data scientists. PerfXPy gives consideration to ease of use and high performance. PerfXPy mainly optimizes the performance of X86, ARM, RISC-V and heterogeneous acceleration devices, helping algorithm developers to use heterogeneous computing platforms more conveniently and release hardware computing power.
PerfXPy Technical White Paper
DOWNLOAD：Please register an account to get real-time updates.
PerfXPy V1.0 Use Guide
DOWNLOAD：Please register an account to get real-time updates.
PerfXPy V1.0 for PHYTIUM
DOWNLOAD：Please register an account to get real-time updates.
PerfXPy Sample Code
Open Source Project: GitHub