]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 96cfc49)
metal : extend mul_mv_ext to BF16, Q2_K, Q3_K (#20250)
authorPaul Flynn <redacted>
Mon, 9 Mar 2026 14:48:12 +0000 (10:48 -0400)
committerGitHub <redacted>
Mon, 9 Mar 2026 14:48:12 +0000 (16:48 +0200)
commite22cd0aa155f0188eb9c605fd54daf3f2f38177b
tree74dc313fc3d8521344e608ccb460b28515299f6btree
parent96cfc4992c625984d99f3529d30bedab1e1ecf93commit | diff
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 <redacted>
ggml/src/ggml-metal/ggml-metal-ops.cpp diff | blob | history
ggml/src/ggml-metal/ggml-metal.metal diff | blob | history
Packaging of ggml-org/llama.cpp