VUDA: A CUDA-Like Runtime API For Vulkan
An independent open-source developer has started the VUDA project as an effort to provide a CUDA Runtime API interface based on Vulkan.
This library consists of just C++ headers to provide a CUDA-like run-time API for writing GPU accelerated applications.
VUDA is licensed under the MIT license and it's important to note that it's not in any way an official NVIDIA project but is just trying to make it easier to utilize Vulkan for GPU computing for those more accustomed to the semantics of the Compute Unified Device Architecture.
An extremely basic code example and other information on this still very young open-source project can be found via VUDA on GitHub.
This library consists of just C++ headers to provide a CUDA-like run-time API for writing GPU accelerated applications.
VUDA is licensed under the MIT license and it's important to note that it's not in any way an official NVIDIA project but is just trying to make it easier to utilize Vulkan for GPU computing for those more accustomed to the semantics of the Compute Unified Device Architecture.
An extremely basic code example and other information on this still very young open-source project can be found via VUDA on GitHub.
13 Comments