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 12, 14:45 UTC
[XPU] upgrade torch & ipex 2.9 on XPU platform (#29307)
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Kunshang Ji committed
Nov 25, 2025 at 09:34 UTC
b8328b49fb9954575bd1d7b30b22bb626ee47624
2 files changed
+12
-7
docker/Dockerfile.xpu
+9
-4
@@ -1,4 +1,4 @@
1
-FROM intel/deep-learning-essentials:2025.1.3-0-devel-ubuntu24.04 AS vllm-base
1
+FROM intel/deep-learning-essentials:2025.2.2-0-devel-ubuntu24.04 AS vllm-base
2
3
RUN wget -O- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB | gpg --dearmor | tee /usr/share/keyrings/oneapi-archive-keyring.gpg > /dev/null && \
4
echo "deb [signed-by=/usr/share/keyrings/oneapi-archive-keyring.gpg] https://apt.repos.intel.com/oneapi all main" | tee /etc/apt/sources.list.d/oneAPI.list && \
@@ -25,10 +25,14 @@ RUN apt clean && apt-get update -y && \
25
RUN update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.12 1
26
RUN update-alternatives --install /usr/bin/python python /usr/bin/python3.12 1
27
28
-RUN apt install -y libze1 libze-dev libze-intel-gpu1 intel-opencl-icd libze-intel-gpu-raytracing
28
+RUN apt install -y libze1 libze-dev libze-intel-gpu1 intel-opencl-icd libze-intel-gpu-raytracing intel-ocloc
29
+
30
+# This oneccl contains the BMG support which is not the case for default version of oneapi 2025.2.
31
+RUN wget https://github.com/uxlfoundation/oneCCL/releases/download/2021.15.6/intel-oneccl-2021.15.6.9_offline.sh
32
+RUN bash intel-oneccl-2021.15.6.9_offline.sh -a --silent --eula accept && \
33
+ echo "source /opt/intel/oneapi/setvars.sh --force" >> /root/.bashrc && \
34
+ echo "source /opt/intel/oneapi/ccl/2021.15/env/vars.sh --force" >> /root/.bashrc
35
30
-RUN wget https://github.com/uxlfoundation/oneCCL/releases/download/2021.15.4/intel-oneccl-2021.15.4.11_offline.sh
31
-RUN bash intel-oneccl-2021.15.4.11_offline.sh -a --silent --eula accept && echo "source /opt/intel/oneapi/setvars.sh --force" >> /root/.bashrc
36
SHELL ["bash", "-c"]
37
CMD ["bash", "-c", "source /root/.bashrc && exec bash"]
38
@@ -72,6 +76,7 @@ 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
+# remove torch bundled oneccl to avoid conflicts
80
RUN --mount=type=cache,target=/root/.cache/pip \
81
pip uninstall oneccl oneccl-devel -y
82
requirements/xpu.txt
+3
-3
@@ -10,9 +10,9 @@ wheel
10
jinja2>=3.1.6
11
datasets # for benchmark scripts
12
numba == 0.61.2 # Required for N-gram speculative decoding
13
-torch==2.8.0+xpu
13
+--extra-index-url=https://download.pytorch.org/whl/xpu
14
+torch==2.9.0+xpu
15
torchaudio
16
torchvision
16
---extra-index-url=https://download.pytorch.org/whl/xpu
17
18
-intel-extension-for-pytorch @ https://intel-extension-for-pytorch.s3.us-east-1.amazonaws.com/ipex_dev/xpu/intel_extension_for_pytorch-2.8.10.post1%2Bxpu-cp312-cp312-linux_x86_64.whl
18
+intel-extension-for-pytorch @ https://intel-extension-for-pytorch.s3.us-east-1.amazonaws.com/ipex_dev/xpu/intel_extension_for_pytorch-2.9.10.post0%2Bxpu-cp312-cp312-linux_x86_64.whl