# Hugging Face TRL fix halves a Qwen3 H100 step in SDPA test

> Source: <https://runtimewire.com/article/hugging-face-trl-packed-batches-rows-h100-trainium>
> Published: 2026-11-14 05:41:34+00:00

# Hugging Face TRL fix halves a Qwen3 H100 step in SDPA test

**Hugging Face contributor tests report an H100 SDPA step falling from 2.06 to 1.02 seconds; on Trainium2, the revised code produced step timings where the previous path crashed.**

        By [RuntimeWire Staff](https://runtimewire.com/author/runtimewire-staff)
        · Published 

Hugging Face, the open-source AI company co-founded by Clement Delangue, Julien Chaumond and Thomas Wolf, merged a TRL fix on November 13th, 2026 that halves a reported Qwen3 training step on an H100 and produced step timings on AWS Trainium2 where the previous layout crashed. The merged pull request keeps best-fit-decreasing packed examples as separate batch rows for standard attention models, instead of flattening each packed batch into one long row by default.…