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.
And lastly, no mention is made of GPU or TPU inference. This is Tensorflow's killer feature, and where its performance per compute dollar gets a lot better. I am very eager to see projects like rust-gpu bring us to the promised land of writing GPU compute programs in Rust and remove Tensorflow's advantage in this area, but it's still a little while away.
I have used the https://github.com/nbigaouette/onnxruntime-rs ONNX C++ wrapper on a Pytorch model, and did not see any difference in compute time between ONNX Python and ONNX Rust for GPU.
NOTE:
The number of mentions on this list indicates mentions on common posts plus user suggested alternatives.
Hence, a higher number means a more popular project.
Related posts
- What's the coolest Rust project you've seen that made you go, 'Wow, I didn't know Rust could do that!'?
- [rust-gpu] How do I run/build my own shaders locally?
- Posh: Type-Safe Graphics Programming in Rust
- Introducing posh: Type-Safe Graphics Programming in Rust
- Looking for high level GPU computing crate