[Hardware][CPU] Update intel_extension_for_pytorch 2.7.0 and move to requirements/cpu.txt

Signed-off-by: Kay Yan <[email protected]>

Author: yankay

docker/Dockerfile.cpu

@@ -51,9 +51,6 @@ RUN --mount=type=cache,target=/root/.cache/uv \
     uv pip install --upgrade pip && \
     uv pip install -r requirements/cpu.txt
 
-RUN --mount=type=cache,target=/root/.cache/uv \
-    uv pip install intel-openmp==2024.2.1 intel_extension_for_pytorch==2.6.0
-
 ENV LD_PRELOAD="/usr/lib/x86_64-linux-gnu/libtcmalloc_minimal.so.4:/opt/venv/lib/libiomp5.so:$LD_PRELOAD"
 
 RUN echo 'ulimit -c 0' >> ~/.bashrc

requirements/cpu.txt

@@ -20,3 +20,7 @@ datasets # for benchmark scripts
 
 # cpu cannot use triton 3.3.0
 triton==3.2.0; platform_machine == "x86_64"
+
+# Intel Extension for PyTorch, only for x86_64 CPUs
+intel-openmp; platform_machine == "x86_64"
+intel_extension_for_pytorch==2.7.0; platform_machine == "x86_64"

vllm/model_executor/layers/quantization/ipex_quant.py

@@ -14,7 +14,7 @@
 from vllm.model_executor.layers.quantization.gptq import GPTQLinearMethod
 from vllm.platforms import current_platform
 
-MIN_IPEX_VERSION = "2.5.0"
+MIN_IPEX_VERSION = "2.7.0"
 
 
 class IPEXConfig(QuantizationConfig):