LodeHQ
AI news for people who shipFree, by email

AI · pytorch.org · Sunday, 15 November 2026

Helion cuts kernel tuning with compile-time performance guesses

Helion can now choose kernel settings at compile time, so developers avoid the long autotuning runs that previously took up to 25 hours for one paged-attention kernel across 72 shapes. It reads the kernel’s intermediate representation and dataflow, classifies it as matmul, reduction, or pointwise, then selects block sizes, warps, and stages. On H100, the approach was 2.13x faster than Helion’s old no-autotune default for linear attention and 2.2x faster across six vLLM reduction kernels.

Read the source: pytorch.org ↗In the 15 November issue →