The PyTorch Foundation is welcoming two new projects as it expands oversight of fundamental open source tools in the AI stack, including a newly donated Meta project.

Helion, a Meta-built domain-specific language (DSL) for writing machine learning kernels, and Hugging Face’s tensor serialization format Safetensors have both joined the foundation this week.

PyTorch Foundation CTO Matt White told The Stack both projects support the organisation’s aim to offer the different components “needed to complete the entire AI life cycle.”

He said: “Having portable formats that work across different frameworks is extremely important to be able to ship and move models around. And then Helion makes things more accessible for folks that want to do custom kernel development.”

Get the full story: Subscribe for free

Join peers managing over $100 billion in annual IT spend and subscribe to unlock full access to The Stack’s analysis and events.

Subscribe now

Already a member? Sign in