Automating CUDA-to-OpenCL Translation

  • If I have seen further it is by standing on the shoulders of giants — Issac Newton

About CU2CL

The CU2CL translator project seeks to deliver platform portability of CUDA codes via robust automated translation to OpenCL.


You already wrote your code for one platform, let CU2CL automatically write your code for the rest of the platforms.

The community has invested significantly in the development of highly-accelerated CUDA codes. However this software is currently vendor-specific, requiring lengthy manual translation efforts to make use of alternative hardware platforms. By providing automatic translation of these codes to a vendor-agnostic and open-standard computing language, we significantly enhance the productivity of developers and users to execute their codes across a much broader range of hardware platforms than ever before.


CUDA developers and users who wish their code was portable to more accelerators from other vendors.

Getting Started

Go to our documentation, which includes release notes, installation instructions, tutorial, and FAQ.


Xilinx Harris

As a thanks for supporting our work, sponsors receive early access to all major releases. If you are interested in becoming a sponsor, please contact us.