Closed gusanmaz closed 11 months ago
I believe llama.cpp has some CUDA support. I include llama.cpp as a sub repo so I would recommend looking at their docs to see whether this is true and if so, whether any build flags need to be enabled.
I am able to run llava-cpp-server on CPU without an issue. Is there a way to run llava-cpp-server on GPU?
Hi gusanmaz, what speed can you run on cpu?
I am able to run llava-cpp-server on CPU without an issue. Is there a way to run llava-cpp-server on GPU?