]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
metal : optimize FA vec for large sequences and BS <= 8 (llama/15566)
authorGeorgi Gerganov <redacted>
Tue, 26 Aug 2025 11:22:14 +0000 (14:22 +0300)
committerGeorgi Gerganov <redacted>
Sat, 20 Sep 2025 10:42:42 +0000 (13:42 +0300)
commit1c21a850bea65802f276197a35f5506ee33020b4
tree2d78888a602372c4855f7cf5568dd540bea10f69
parentdc693ca8c96ed8ea00b60dce92d7344d92f99aad
metal : optimize FA vec for large sequences and BS <= 8 (llama/15566)

* metal : optmize FA vec for large heads and sequences

* metal : adjust small-batch mul mv kernels

ggml-ci

* batched-bench : fix total speed computation

ggml-ci

* cont : add comments

ggml-ci
ggml/src/ggml-metal/ggml-metal-impl.h
ggml/src/ggml-metal/ggml-metal.m
ggml/src/ggml-metal/ggml-metal.metal