]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
metal : FA support F32 K and V and head size = 32 (#16531)
authorGeorgi Gerganov <redacted>
Mon, 13 Oct 2025 20:07:57 +0000 (23:07 +0300)
committerGitHub <redacted>
Mon, 13 Oct 2025 20:07:57 +0000 (23:07 +0300)
commite60f241eacec42d3bd7c9edd37d236ebf35132a8
treee7520545b0ccac6ab9d0986b61ae5020d091d39e
parente38b7c6e9e4453e3b3e96d76e38bc2ccb6bce458
metal : FA support F32 K and V and head size = 32 (#16531)

* metal : FA support F32 K and V and head size = 32

* graph : remove obsolete comment [no ci]
ggml/src/ggml-metal/ggml-metal-device.m
ggml/src/ggml-metal/ggml-metal.metal
src/llama-graph.cpp
tests/test-backend-ops.cpp