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 13, 02:00 UTC
Run macos smoke test workflow on main commit (#28752)
Signed-off-by: Michael Goin <mgoin64@gmail.com> Signed-off-by: mgoin <mgoin64@gmail.com>
Michael Goin committed
Nov 17, 2025 at 22:16 UTC
88ab591f0b20c28cb167fd65d10ccade99d873ae
2 files changed
+13
-7
.github/workflows/macos-smoke-test.yml
+10
-5
@@ -1,6 +1,9 @@
1
name: macOS Apple Silicon Smoke Test
2
3
on:
4
+ push:
5
+ branches:
6
+ - main
7
workflow_dispatch: # Manual trigger
8
9
jobs:
@@ -19,13 +22,15 @@ jobs:
22
pyproject.toml
23
python-version: '3.12'
24
22
- - name: Install dependencies
25
+ - name: Create virtual environment
26
run: |
24
- uv pip install -r requirements/cpu-build.txt
25
- uv pip install -r requirements/cpu.txt
27
+ uv venv
28
+ echo "$GITHUB_WORKSPACE/.venv/bin" >> "$GITHUB_PATH"
29
27
- - name: Build vLLM
28
- run: uv pip install -v -e .
30
+ - name: Install dependencies and build vLLM
31
+ run: |
32
+ uv pip install -r requirements/cpu.txt --index-strategy unsafe-best-match
33
+ uv pip install -e .
34
env:
35
CMAKE_BUILD_PARALLEL_LEVEL: 4
36
requirements/cpu-build.txt
+3
-2
@@ -4,8 +4,9 @@ packaging>=24.2
4
setuptools>=77.0.3,<81.0.0
5
setuptools-scm>=8
6
--extra-index-url https://download.pytorch.org/whl/cpu
7
-torch==2.8.0+cpu; platform_machine == "x86_64"
8
-torch==2.8.0; platform_machine == "ppc64le" or platform_machine == "aarch64" or platform_system == "Darwin"
7
+torch==2.8.0+cpu; platform_machine == "x86_64" or platform_machine == "s390x"
8
+torch==2.9.0; platform_system == "Darwin"
9
+torch==2.8.0; platform_machine == "ppc64le" or platform_machine == "aarch64"
10
scons; platform_machine == "aarch64" # needed to build Arm Compute Library (ACL)
11
wheel
12
jinja2>=3.1.6