llama.cpp
CUDA: build archs as virtual for GGML_NATIVE=OFF
#13135

Merged

CUDA: build archs as virtual for GGML_NATIVE=OFF #13135

JohannesGaessler merged 1 commit into ggml-org:master from JohannesGaessler:cuda-jit

github-actions added Nvidia GPU

github-actions added ggml

JohannesGaessler force pushed 224 days ago

slaren approved these changes on 2025-05-06

CUDA: mix virt/real CUDA archs for GGML_NATIVE=OFF

d9330f75

JohannesGaessler force pushed to d9330f75 224 days ago

JohannesGaessler merged 141a908a into master 224 days ago

Reviewers

slaren

Assignees

No one assigned

Labels

Nvidia GPU ggml

Milestone

No milestone