The paper presents a package of 38 Fortran subprograms for basic numerical linear algebra operations, known as the Basic Linear Algebra Subprograms (BLAS). These subprograms are designed to be used in Fortran and include operations such as dot product, vector operations, Givens transformations, vector norms, scaling, and finding the index of the largest vector component. The subprograms are available in portable Fortran and in assembly language for specific computers like IBM 360/67, CDC 6600, and Univac 1108. The package was developed through a collaborative project and aims to provide standardized, efficient, and portable routines for linear algebra operations. The subprograms are designed to be modular, improving code readability and efficiency. They are used in various numerical linear algebra algorithms, including Gaussian elimination and orthogonal transformations. The package includes functions for single, double, and complex data types. The subprograms are tested on multiple machines to ensure portability and efficiency. The paper also discusses the implementation details, testing procedures, and performance results for different machines. The BLAS package is intended to serve as a foundation for higher-level linear algebra operations and is widely used in scientific computing.The paper presents a package of 38 Fortran subprograms for basic numerical linear algebra operations, known as the Basic Linear Algebra Subprograms (BLAS). These subprograms are designed to be used in Fortran and include operations such as dot product, vector operations, Givens transformations, vector norms, scaling, and finding the index of the largest vector component. The subprograms are available in portable Fortran and in assembly language for specific computers like IBM 360/67, CDC 6600, and Univac 1108. The package was developed through a collaborative project and aims to provide standardized, efficient, and portable routines for linear algebra operations. The subprograms are designed to be modular, improving code readability and efficiency. They are used in various numerical linear algebra algorithms, including Gaussian elimination and orthogonal transformations. The package includes functions for single, double, and complex data types. The subprograms are tested on multiple machines to ensure portability and efficiency. The paper also discusses the implementation details, testing procedures, and performance results for different machines. The BLAS package is intended to serve as a foundation for higher-level linear algebra operations and is widely used in scientific computing.