CUTLASS is described in the following documents and the accompanying Doxygen documentation.

• Quick Start Guide - build and run CUTLASS
• Functionality - summarizes functionality available in CUTLASS
• Efficient GEMM in CUDA - describes how GEMM kernels may be implemented efficiently in CUDA
• GEMM API - describes the CUTLASS GEMM model and C++ template concepts
• Implicit GEMM Convolution - describes 2-D and 3-D convolution in CUTLASS
• Code Organization - describes the organization and contents of the CUTLASS project
• Terminology - describes terms used in the code
• Programming Guidelines - guidelines for writing efficient modern CUDA C++
• Fundamental types - describes basic C++ classes used in CUTLASS to represent numeric quantities and arrays
• Layouts - describes layouts of matrices and tensors in memory
• Tile Iterators - describes C++ concepts for iterating over tiles of matrices in memory
• CUTLASS Profiler - command-line driven profiling application
• CUTLASS Utilities - additional templates used to facilitate rapid development