Helion Inductor Integration

compiler
1

I am curious about the plan for Helion support in Inductor. Is there an RFC for this?

For a given graph, is the intention to have the backend be purely triton vs purely Helion or is there a strategy to be able to mix and match. For mixing and matching, is there a discussion about mitigating fusion barriers?

2

We’re currently working on Helion + Inductor fusion. Please see RFC here: [RFC] Helion kernel + Inductor ops prologue / epilogue fusion · Issue #1346 · pytorch/helion · GitHub

3

Thank you, this is exactly what I was looking for!