# PyTorch blog on "From Minutes to Seconds: LLM-Guided Autotuning for Helion Kernels" **URL:** \ **Category:** Helion\ **Created:** [June 28, 2026, 2:02am UTC](https://dev-discuss.pytorch.org/t/pytorch-blog-on-from-minutes-to-seconds-llm-guided-autotuning-for-helion-kernels/3407 "2026-06-28T02:02:16Z")\ **Posts on this page:** 1\ **Page:** 1
**Author:** ![jongsok.choi](https://avatars.discourse-cdn.com/v4/letter/j/e495f1/32.png) [@jongsok.choi](https://dev-discuss.pytorch.org/u/jongsok.choi)\ **Post date:** [June 28, 2026, 2:02am UTC](https://dev-discuss.pytorch.org/t/pytorch-blog-on-from-minutes-to-seconds-llm-guided-autotuning-for-helion-kernels/3407/1 "2026-06-28T02:02:16Z")
> **[From Minutes to Seconds: LLM-Guided Autotuning for Helion Kernels – PyTorch](https://pytorch.org/blog/from-minutes-to-seconds-llm-guided-autotuning-for-helion-kernels/)** Abstract: Helion, PyTorch’s domain-specific language (DSL) for performance portable machine learning kernels, heavily relies on autotuning for performance. Currently Helion searches utilize the Likelihood-Free Bayesian Optimization (LFBO) to find the most performant configs. LFBO is a strong baseline which works well, but it still grinds through hundreds of compile-and-benchmark cycles per kernel. To this end, we introduce an LLM-guided autotuner that matches LFBO-level kernel performance (geomean 1.009X) while benchmarking ~10X fewer configurations in ~6.7X less wall-clock time. For the handful of kernels where the LLM trails by \>5%, a hybrid strategy (LLM seeding followed by LFBO refinement) closes the gap while remaining ~3X cheaper than the full LFBO search. Finally, the result is largely LLM model-independent — Opus-4.8, gpt-5.5, and Sonnet-4.6 perform within a couple percent of each other — showing that LLM-guided autotuning is a practical approach to dramatically faster kernel tuning at production quality.