forked from vllm-project/vllm
-
Notifications
You must be signed in to change notification settings - Fork 0
81 lines (68 loc) · 2.1 KB
/
macos-smoke-test.yml
File metadata and controls
81 lines (68 loc) · 2.1 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
name: macOS Apple Silicon Smoke Test
on:
push:
branches:
- main
workflow_dispatch: # Manual trigger
jobs:
macos-m1-smoke-test:
runs-on: macos-latest
timeout-minutes: 20
steps:
- uses: actions/checkout@v4
- uses: astral-sh/setup-uv@v7
with:
enable-cache: true
cache-dependency-glob: |
requirements/**/*.txt
pyproject.toml
python-version: '3.12'
- name: Create virtual environment
run: |
uv venv
echo "$GITHUB_WORKSPACE/.venv/bin" >> "$GITHUB_PATH"
- name: Install dependencies and build vLLM
run: |
uv pip install -r requirements/cpu.txt --index-strategy unsafe-best-match
uv pip install -e .
env:
CMAKE_BUILD_PARALLEL_LEVEL: 4
- name: Verify installation
run: |
python -c "import vllm; print(f'vLLM version: {vllm.__version__}')"
python -c "import torch; print(f'PyTorch: {torch.__version__}')"
- name: Smoke test vllm serve
timeout-minutes: 10
run: |
# Start server in background
vllm serve Qwen/Qwen3-0.6B \
--max-model-len=2048 \
--load-format=dummy \
--enforce-eager \
--port 8000 &
SERVER_PID=$!
# Wait for server to start
for i in {1..30}; do
if curl -s http://localhost:8000/health > /dev/null; then
echo "Server started successfully"
break
fi
if [ "$i" -eq 30 ]; then
echo "Server failed to start"
kill "$SERVER_PID"
exit 1
fi
sleep 2
done
# Test health endpoint
curl -f http://localhost:8000/health
# Test completion
curl -f http://localhost:8000/v1/completions \
-H "Content-Type: application/json" \
-d '{
"model": "Qwen/Qwen3-0.6B",
"prompt": "Hello",
"max_tokens": 5
}'
# Cleanup
kill "$SERVER_PID"