EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

[Build] Fix x86 image build (#327)

Browse Source 
Install cpu version of pytorch in x86 to reduce image size

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
wangxiyuan
2025-03-14 09:41:57 +08:00
committed by GitHub
parent c25631ec7b
commit ac1ba1d8d2
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
Dockerfile
View File

@@ -37,7 +37,7 @@ RUN pip config set global.index-url ${PIP_INDEX_URL}
ARG VLLM_REPO=https://github.com/vllm-project/vllm.git
ARG VLLM_TAG=main
RUN git clone --depth 1 $VLLM_REPO --branch $VLLM_TAG /workspace/vllm
RUN VLLM_TARGET_DEVICE="empty" python3 -m pip install /workspace/vllm/
RUN VLLM_TARGET_DEVICE="empty" python3 -m pip install /workspace/vllm/ --extra-index https://download.pytorch.org/whl/cpu/
# In x86, triton will be installed by vllm. But in Ascend, triton doesn't work correctly. we need to uninstall it.
RUN python3 -m pip uninstall -y triton