Triton - Open source neural network programming on GPUs

[igor_kavinski]

Introducing Triton: Open-source GPU programming for neural networks

We’re releasing Triton 1.0, an open-source Python-like programming language which enables researchers with no CUDA experience to write highly efficient GPU code—most of the time on par with what an expert would be able to produce.

openai.com

Hopefully CUDA dies soon? 🙂

 

[lantis3]

https://github.com/openai/triton Triton 2.0 is out.

Another AI language.

Mojo 🔥: Powerful CPU+GPU Programming

Mojo is a programming language that unifies high-level AI development with low-level systems programming. Write once, deploy everywhere - from CPUs to GPUs - without vendor lock-in.

www.modular.com

A Python superset, claims faster than Python 35,000x

Mojo FAQ | Modular

Answers to questions we expect about Mojo.

docs.modular.com

Seems not open source yet.

GitHub - modular/modular: The Modular Platform (includes MAX & Mojo)

The Modular Platform (includes MAX & Mojo). Contribute to modular/modular development by creating an account on GitHub.

github.com

Never a good programmer, and I'm too old to learn new languages. 🙄