minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[diffusion][CI]: Add individual component accuracy CI for diffusion models (#18709)

Browse Source 
Co-authored-by: Xiaoyu Zhang <35585791+BBuf@users.noreply.github.com>
This commit is contained in:
Ratish P
2026-04-01 21:51:36 +08:00
committed by GitHub
co-authored by Xiaoyu Zhang
parent e67b95d66b
commit 4f5b55e379
13 changed files with 2670 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.github/workflows/pr-test-multimodal-gen.yml
+102
View File
@@ -156,6 +156,108 @@ jobs:
with:
artifact-suffix: ${{ matrix.part }}
multimodal-gen-component-accuracy-1-gpu:
if: |
(inputs.target_stage == 'multimodal-gen-component-accuracy-1-gpu') ||
(
!inputs.target_stage &&
((github.event_name == 'schedule' || inputs.test_parallel_dispatch == 'true') || (inputs.caller_needs_failure != 'true' && !cancelled())) &&
inputs.multimodal_gen == 'true'
)
runs-on: 1-gpu-h100
timeout-minutes: 240
steps:
- name: Checkout code
uses: actions/checkout@v4
with:
ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
- uses: ./.github/actions/check-stage-health
- uses: ./.github/actions/check-maintenance
- name: Download artifacts
if: inputs.sgl_kernel == 'true'
uses: actions/download-artifact@v4
with:
path: sgl-kernel/dist/
merge-multiple: true
pattern: wheel-python3.10-cuda12.9
- name: Install dependencies
timeout-minutes: 20
run: |
CUSTOM_BUILD_SGL_KERNEL=${{inputs.sgl_kernel}} bash scripts/ci/cuda/ci_install_dependency.sh diffusion
- name: Run diffusion component accuracy tests (1-GPU)
timeout-minutes: 240
env:
RUNAI_STREAMER_MEMORY_LIMIT: 0
run: |
cd python
if [ "${{ inputs.continue_on_error }}" = "true" ]; then
exit_code=0
python3 -m pytest -s -v sglang/multimodal_gen/test/server/test_accuracy_1_gpu_a.py || exit_code=$?
python3 -m pytest -s -v sglang/multimodal_gen/test/server/test_accuracy_1_gpu_b.py || exit_code=$?
exit $exit_code
fi
python3 -m pytest -s -v sglang/multimodal_gen/test/server/test_accuracy_1_gpu_a.py
python3 -m pytest -s -v sglang/multimodal_gen/test/server/test_accuracy_1_gpu_b.py
- uses: ./.github/actions/upload-cuda-coredumps
if: always()
multimodal-gen-component-accuracy-2-gpu:
if: |
(inputs.target_stage == 'multimodal-gen-component-accuracy-2-gpu') ||
(
!inputs.target_stage &&
((github.event_name == 'schedule' || inputs.test_parallel_dispatch == 'true') || (inputs.caller_needs_failure != 'true' && !cancelled())) &&
inputs.multimodal_gen == 'true'
)
runs-on: 2-gpu-h100
timeout-minutes: 240
steps:
- name: Checkout code
uses: actions/checkout@v4
with:
ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
- uses: ./.github/actions/check-stage-health
- uses: ./.github/actions/check-maintenance
- name: Download artifacts
if: inputs.sgl_kernel == 'true'
uses: actions/download-artifact@v4
with:
path: sgl-kernel/dist/
merge-multiple: true
pattern: wheel-python3.10-cuda12.9
- name: Install dependencies
timeout-minutes: 20
run: |
CUSTOM_BUILD_SGL_KERNEL=${{inputs.sgl_kernel}} bash scripts/ci/cuda/ci_install_dependency.sh diffusion
- name: Run diffusion component accuracy tests (2-GPU)
timeout-minutes: 240
env:
RUNAI_STREAMER_MEMORY_LIMIT: 0
run: |
cd python
if [ "${{ inputs.continue_on_error }}" = "true" ]; then
exit_code=0
torchrun --nproc_per_node=2 -m pytest -s -v sglang/multimodal_gen/test/server/test_accuracy_2_gpu_a.py || exit_code=$?
torchrun --nproc_per_node=2 -m pytest -s -v sglang/multimodal_gen/test/server/test_accuracy_2_gpu_b.py || exit_code=$?
exit $exit_code
fi
torchrun --nproc_per_node=2 -m pytest -s -v sglang/multimodal_gen/test/server/test_accuracy_2_gpu_a.py
torchrun --nproc_per_node=2 -m pytest -s -v sglang/multimodal_gen/test/server/test_accuracy_2_gpu_b.py
- uses: ./.github/actions/upload-cuda-coredumps
if: always()
multimodal-gen-unit-test:
if: |
(inputs.target_stage == 'multimodal-gen-unit-test') ||
.github/workflows/pr-test.yml
+2
View File
@@ -890,6 +890,8 @@ jobs:
(
inputs.target_stage == 'multimodal-gen-test-1-gpu' ||
inputs.target_stage == 'multimodal-gen-test-2-gpu' ||
inputs.target_stage == 'multimodal-gen-component-accuracy-1-gpu' ||
inputs.target_stage == 'multimodal-gen-component-accuracy-2-gpu' ||
inputs.target_stage == 'multimodal-gen-unit-test' ||
(
!inputs.target_stage &&