A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 11, 11:15 UTC
[XPU] fix Dockerfile.xpu, avoid wheel conflicts (#30662)
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Kunshang Ji committed
Dec 15, 2025 at 13:32 UTC
e3a1cd1c59b7cfb8fd6eb05e69393aa7f42dc12d
1 file changed
+3
docker/Dockerfile.xpu
+3
@@ -76,6 +76,9 @@ RUN python3 -m pip install -e tests/vllm_test_utils
76
ENV NIXL_VERSION=0.7.0
77
RUN python3 /workspace/vllm/tools/install_nixl_from_source_ubuntu.py
78
79
+# PyJWT-2.7.0 will influence some wheel behaviors, remove its dist-info to avoid conflicts
80
+RUN rm /usr/lib/python3/dist-packages/PyJWT-2.7.0.dist-info/ -rf
81
+
82
# remove torch bundled oneccl to avoid conflicts
83
RUN --mount=type=cache,target=/root/.cache/pip \
84
pip uninstall oneccl oneccl-devel -y