adaptivecpp ☆
« Back to VersTracker
Description:
SYCL and C++ standard parallelism for CPUs and GPUs
Type: Formula  |  Latest Version: 25.10.0@0  |  Tracked Since: Dec 17, 2025
Links: Homepage  |  formulae.brew.sh
Category: Developer tools
Tags: sycl c++ hpc gpu open-source compiler
Install: brew install adaptivecpp
About:
AdaptiveCpp is a compiler for C++ standard parallelism and SYCL, enabling high-performance code execution across diverse hardware including CPUs and GPUs from various vendors. It provides a unified, open-source toolchain that allows developers to write portable parallel code without vendor-specific lock-in.
Key Features:
  • Support for SYCL and C++ standard parallelism (std::execution::par)
  • Multi-vendor hardware support (NVIDIA, AMD, Intel, CPUs)
  • Ahead-of-time (AOT) and just-in-time (JIT) compilation
  • Open-source and vendor-agnostic
Use Cases:
  • Porting HPC applications to run on heterogeneous clusters
  • Accelerating C++ compute workloads on consumer GPUs
  • Developing vendor-neutral parallel algorithms
Alternatives:
  • Intel oneAPI DPC++ – Intel's official SYCL implementation, focused primarily on Intel hardware, whereas AdaptiveCpp offers broader multi-vendor support.
  • CUDA / ROCm – Vendor-specific platforms requiring code rewrite; AdaptiveCpp allows standard C++ code to run on these platforms portably.
License: BSD-2-Clause
Dependencies: lld, llvm, numactl
Bottles available for: arm64_tahoe, arm64_sequoia, arm64_sonoma, sonoma, arm64_linux, x86_64_linux
Version History
Detected Version Rev Change Commit
Nov 6, 2025 12:11am 0 VERSION_BUMP 6765fe05
Sep 12, 2025 6:34am 1 VERSION_BUMP 6c320347
« Back to VersTracker  |  Browse Packages  |  About