]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 6b36c23)
ggml-cuda: Allow transpose-free gemmv computation (#26171)
authorRobert Esclapez <redacted>
Thu, 30 Jul 2026 13:39:46 +0000 (15:39 +0200)
committerGitHub <redacted>
Thu, 30 Jul 2026 13:39:46 +0000 (21:39 +0800)
commite1a1abb78746c025f5e9039f590e37ccdb758ae7
tree953f0eca60b124146cce5bc581b03aea248ee8f4tree
parent6b36c2305644fd30db7cce3f4840c74574f31ce9commit | diff
ggml-cuda: Allow transpose-free gemmv computation (#26171)

When matrix's weights are shaped 1xK is leverage a transpose-free
computation to use mat_mul_vec_f.
ggml/src/ggml-cuda/ggml-cuda.cu diff | blob | history
tests/test-backend-ops.cpp diff | blob | history
Packaging of ggml-org/llama.cpp