Skip to content

Lab Notes6 field reports

All Lab Notes

Tag archive

#gptq

GPTQ · 1 field report

Filed under Local AI (1). All tags.

Chronological feed

  1. Local AI · 10 min read

    Squeezing Qwen3.8 Until It Cried: A 46.9% Speedup and One Optimization Too Far

    A 46.9% Qwen3.8 speedup, a production qualification, and the tempting 64K shortcut that failed when the benchmark got serious.

    Promote the full-vocabulary checkpoint: 46.88% faster decode, qualified behavior, and no support for the later 64K shortcut.