#706 Feat/q2 avx2 bitplane

closed · @merbanan · 0 comments · View on GitHub

Description

A new AVX2 row kernel for Q2_0 experts on CPUs without AVX-512, the default path in `q2_0_gguf_rows_multi_avx2`.

**Single thread, weights in L2:** 2.3–2.8× faster at every token count from 1 to 4.

Related on strata.com

Install Strata
Model sizes & RAM
Release notes

Editorial links to help you install, pick models, or read release notes — not part of the upstream thread.