Show HN: Automatic deduction of FlashAttention kernel from ~25 lines of SQL
By alxmrs · 2026-10-30 · 11 points · 1 comments
https://github.com/xqlsystems/xarray-sql/blob/6debe2495dcb4e8fac20ab97df5b89baae2a31a6/benchmarks/ml/01_attention.py
In this demo, we automatically deduce a FlashAttention kernel with ~25 lines of SQL, which just expresses vanilla attention. This is possible due to three portable SQL extensions that we’re building: - xarray-sql: a library that efficiently represents petabyte scale array datase…
Open the full discussion on BetterNews