]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 48e2fa9)
CUDA: use fastdiv + ggml_cuda_mad for mmvf (#16557)
authorAman Gupta <redacted>
Tue, 14 Oct 2025 11:16:21 +0000 (19:16 +0800)
committerGitHub <redacted>
Tue, 14 Oct 2025 11:16:21 +0000 (13:16 +0200)
commit1ee9d0b415cdf5240418c110a18b419f4002b154
tree87759f7fce7749da3a2119720937df0307779382tree
parent48e2fa9fb7c2de1e53808fdb65ec33f916020fc4commit | diff
CUDA: use fastdiv + ggml_cuda_mad for mmvf (#16557)

* CUDA: use fastdiv + ggml_cuda_mad for mmvf

* use bf16 directly + fix formatting

* Add exception for HIP code
ggml/src/ggml-cuda/mmvf.cu diff | blob | history
Packaging of ggml-org/llama.cpp