Skip to content

Merge pull request #1 from EvolvingLMMs-Lab/dev/fix_multi_img_ci #4

Merge pull request #1 from EvolvingLMMs-Lab/dev/fix_multi_img_ci

Merge pull request #1 from EvolvingLMMs-Lab/dev/fix_multi_img_ci #4

Workflow file for this run

name: MoE Test
on:
push:
branches: [ main ]
paths:
- "python/sglang/**"
- "test/**"
pull_request:
branches: [ main ]
paths:
- "python/sglang/**"
- "test/**"
workflow_dispatch:
concurrency:
group: moe-test-${{ github.ref }}
cancel-in-progress: true
jobs:
moe-test:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 2-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[all]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
- name: Benchmark MoE Serving Throughput
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_moe_serving_throughput.TestServingThroughput.test_default
- name: Benchmark MoE Serving Throughput (w/o RadixAttention)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_moe_serving_throughput.TestServingThroughput.test_default_without_radix_cache