Ananth Veluvali(@AnanthVeluvali):I connected devin to some @modal H100s and then left it alone to run its own experiment loop against the GPUs. it found a way to reduce peak memory by up to 46% and latency by up to 52% on a popular OS training-kernel repo. since the triton kernel was already near peak HBM bandwidth, it focused on other things, removing a 2GB allocation that was happening on every chunk of the backward pass. then it ran the benchmarks to confirm the speedup 🤯

Ananth Veluvali

@AnanthVeluvali

ars longa, vita brevis | prev. exited founder, @stanford

가입 April 2022

299 팔로잉 중 1.1K 팬

Ananth Veluvali@AnanthVeluvali

2026.09.14 04:05

I connected devin to some @modal H100s and then left it alone to run its own experiment loop against the GPUs. it found a way to reduce peak memory by up to 46% and latency by up to 52% on a popular OS training-kernel repo. since the triton kernel was already near peak HBM bandwidth, it focused on other things, removing a 2GB allocation that was happening on every chunk of the backward pass. then it ran the benchmarks to confirm the speedup 🤯