diff --git a/server/Makefile b/server/Makefile index 0db6f89b..3abc917e 100644 --- a/server/Makefile +++ b/server/Makefile @@ -39,6 +39,7 @@ install: install-cuda install-cuda: install-server install-flash-attention-v2-cuda install-flash-attention uv pip install -e ".[attention,bnb,marlin,moe]" uv pip install nvidia-nccl-cu12==2.22.3 + kernels download . install-rocm: install-server install-flash-attention-v2-rocm install-vllm-rocm