all options
trixie  ] [  sid  ] [  experimental  ]
[ Source: nvidia-cuda-toolkit  ]

Package: libcublaslt12 (12.2.1.16~12.2.0-1 and others) [non-free]

Links for libcublaslt12

Screenshot

Debian Resources:

Download Source Package nvidia-cuda-toolkit:

Maintainers:

External Resources:

Similar packages:

Experimental package

Warning: This package is from the experimental distribution. That means it is likely unstable or buggy, and it may even cause data loss. Please be sure to consult the changelog and other possible documentation before using it.

NVIDIA cuBLASLt Library

The Compute Unified Device Architecture (CUDA) enables NVIDIA graphics processing units (GPUs) to be used for massively parallel general purpose computation.

The cuBLASLt library is a lightweight GEMM library with a flexible API and tensor core support for INT8 inputs and FP16 CGEMM split-complex matrix multiplication.

This package contains the cuBLASLt runtime library.

Other Packages Related to libcublaslt12

  • depends
  • recommends
  • suggests
  • enhances

Download libcublaslt12

Download for all available architectures
Architecture Version Package Size Installed Size Files
amd64 12.2.1.16~12.2.0-1+b1 188,572.9 kB503,121.0 kB [list of files]
arm64 12.2.1.16~12.2.0-1+b1 189,413.2 kB481,229.0 kB [list of files]
ppc64el 12.2.1.16~12.2.0-1+b1 140,839.6 kB344,569.0 kB [list of files]