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 17, 00:30 UTC
[Bugfix] Fix a bug caused by pip install setuptools>=49.4.0 for CPU backend (#5249)
Jie Fu (傅杰) committed
Jun 5, 2024 at 00:57 UTC
87d5abef756e959fd05dcaa6d5c00d07c34e1fe8
2 files changed
+2
-2
Dockerfile.cpu
+1
-1
@@ -7,7 +7,7 @@ RUN apt-get update -y \
7
&& update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-12 10 --slave /usr/bin/g++ g++ /usr/bin/g++-12
8
9
RUN pip install --upgrade pip \
10
- && pip install wheel packaging ninja setuptools>=49.4.0 numpy
10
+ && pip install wheel packaging ninja "setuptools>=49.4.0" numpy
11
12
FROM cpu-test-1 AS build
13
docs/source/getting_started/cpu-installation.rst
+1
-1
@@ -54,7 +54,7 @@ Build from source
54
.. code-block:: console
55
56
$ pip install --upgrade pip
57
- $ pip install wheel packaging ninja setuptools>=49.4.0 numpy
57
+ $ pip install wheel packaging ninja "setuptools>=49.4.0" numpy
58
$ pip install -v -r requirements-cpu.txt --extra-index-url https://download.pytorch.org/whl/cpu
59
60
- Finally, build and install vLLM CPU backend: