Our great sponsors
-
WorkOS
The modern identity platform for B2B SaaS. The APIs are flexible and easy-to-use, supporting authentication, user identity, and complex enterprise features like SSO and SCIM provisioning.
If you prefer an open alternative to CUDA, there are complete, easy to use und well documented bindings for opencl: https://github.com/cogciprocate/ocl/
Crates for OpenCL and CUDA APIs are generally lagging behind in feature support. I think we are still long way from seeing v1.0+ releases on those. One of the most active ones in recent development has been opencl3 crate. Then, there are also some examples like blub using compute shaders with wgpu crate.
Crates for OpenCL and CUDA APIs are generally lagging behind in feature support. I think we are still long way from seeing v1.0+ releases on those. One of the most active ones in recent development has been opencl3 crate. Then, there are also some examples like blub using compute shaders with wgpu crate.
Working on autograph, for machine learning and neural networks. Unlike CUDA / HIP it's threadsafe, but doesn't expose low level things like multiple streams. Most of the shaders are glsl but I'm now using rust_gpu for pure rust gpu code.
Working on autograph, for machine learning and neural networks. Unlike CUDA / HIP it's threadsafe, but doesn't expose low level things like multiple streams. Most of the shaders are glsl but I'm now using rust_gpu for pure rust gpu code.
I think at the moment the I would use https://github.com/bheisler/RustaCUDA/ and C/C++ `.cu` files with `extern "C"` to call from Rust. Even though there are experiments to have nvptx as a compilation target for Rust.