llama.cpp
ggml: support CUDA's half type for aarch64(#1455)
#2670

Merged

ggml: support CUDA's half type for aarch64(#1455) #2670

ggerganov merged 2 commits into ggml-org:master from KyL0N:ggml-patch

ggml: support CUDA's half type for aarch64(#1455)

8a25bd41

ggml: use __CUDACC__ to recognise nvcc compiler

16ab5f1b

KyL0N force pushed from 972d6380 to 16ab5f1b 2 years ago

slaren approved these changes on 2023-08-20

ggerganov merged 1e3bc523 into master 2 years ago

KyL0N deleted the ggml-patch branch 2 years ago

Reviewers

slaren

Assignees

No one assigned

Labels

None yet

Milestone

No milestone