]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
CUDA: disable cuda graph when using n-cpu-moe (llama/18593)
authorAman Gupta <redacted>
Sun, 4 Jan 2026 17:37:48 +0000 (01:37 +0800)
committerGeorgi Gerganov <redacted>
Wed, 14 Jan 2026 07:11:59 +0000 (09:11 +0200)
commitf0bf5b8cc32e3b6ea5c549e4283e5f3a0ac5005d
treeeb5ad945c98b28514dd89aab1a8126ef410329df
parent88f5765c823f3aaab9ea706b347317b3ca9a49cf
CUDA: disable cuda graph when using n-cpu-moe (llama/18593)

* CUDA: disable cuda graph when using n-cpu-moe

* call ggml_cuda_set_device
ggml/src/ggml-cuda/ggml-cuda.cu