@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / c16eaac500
Fork 0

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 16, 19:30 UTC

[Hardware][Intel CPU] Update torch 2.4.0 for CPU backend (#6931)

Jie Fu (傅杰) committed Aug 2, 2024 at 23:55 UTC c16eaac5001d9e2bfb51c9812ec0c2b9e32b8d25
2 files changed +2 -2
Dockerfile.cpu
+1 -1
@@ -26,7 +26,7 @@ COPY ./ /workspace/vllm
26
27 WORKDIR /workspace/vllm
28
29 -RUN pip install -v -r requirements-cpu.txt --extra-index-url https://download.pytorch.org/whl/test/cpu
29 +RUN pip install -v -r requirements-cpu.txt --extra-index-url https://download.pytorch.org/whl/cpu
30
31 # Support for building with non-AVX512 vLLM: docker build --build-arg VLLM_CPU_DISABLE_AVX512="true" ...
32 ARG VLLM_CPU_DISABLE_AVX512
requirements-cpu.txt
+1 -1
@@ -2,5 +2,5 @@
2 -r requirements-common.txt
3
4 # Dependencies for x86_64 CPUs
5 -torch == 2.4.0; platform_machine != "ppc64le"
5 +torch == 2.4.0+cpu; platform_machine != "ppc64le"
6 torchvision; platform_machine != "ppc64le" # required for the image processor of phi3v, this must be updated alongside torch