ggml-org/llama.cpp b8261

on GitHub

metal : extend mul_mv_ext to BF16, Q2_K, Q3_K (#20250)

Enable mul_mv_ext small-batch kernels (BS 2-8) for BF16, Q2_K,
and Q3_K quantization types. These types previously fell through
to the slower single-row mul_mv path.

BF16 uses the float4 dequantize path (like F16). Q2_K and Q3_K
use the float4x4 K-quant path (like Q4_K/Q5_K/Q6_K).

Co-authored-by: Claude Opus 4.6 noreply@anthropic.com