]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
CUDA: only allocate FA tmp buffer if needed (#18564)
authorJohannes Gäßler <redacted>
Sat, 3 Jan 2026 12:55:53 +0000 (13:55 +0100)
committerGitHub <redacted>
Sat, 3 Jan 2026 12:55:53 +0000 (13:55 +0100)
commit0f2e42ca1d1d025e6c4cb4bffb78da8972dec17f
treed3b1ec21e529c557f43de11dcd4365323a0ec19a
parent9dba9f5352308894bfb8786fcfe7c284168ff8f5
CUDA: only allocate FA tmp buffer if needed (#18564)
ggml/src/ggml-cuda/fattn-common.cuh