ddemidov/vexcl

VexCL is a C++ vector expression template library for OpenCL/CUDA/OpenMP

What it solves

VexCL simplifies the development of GPGPU (General-Purpose computing on Graphics Processing Units) applications using C++. It reduces the amount of boilerplate code typically required when writing programs for OpenCL and CUDA, making it easier to write high-performance computing code for GPUs.

How it works

The library uses vector expression templates to provide an intuitive notation for vector arithmetic, reductions, and sparse matrix-vector products. This allows developers to write mathematical expressions that the library then translates into efficient GPU kernels.

Who it’s for

C++ developers who need to perform high-performance computations on GPUs across different platforms and devices using OpenCL or CUDA.

Highlights

  • Support for both OpenCL and CUDA platforms.
  • Multi-device and multi-platform computation capabilities.
  • MIT license for permissive use.

Related

  • Project
    nuclia/nucliadb

    NucliaDB is a hybrid AI search database that combines vector, full-text, and graph indexes to store and search unstructured data.

  • Project
    AITabby/codexsplit

    A local control center for Codex Desktop that enables the integration of third-party AI models, official GPT account pooling, and intelligent agent routing.

  • Project
    neo4j-labs/create-context-graph

    A CLI scaffolding tool that generates full-stack, domain-specific AI applications with integrated Neo4j knowledge graphs and agent frameworks.

  • Project
    EricLBuehler/candle-vllm

    A high-performance inference and serving platform for local LLMs featuring an OpenAI-compatible API, multi-GPU/multi-node support, and aggressive KV cache compression.