Migrate VLM tests and remove unit-test-backend-1-gpu job (#16679)

.github/workflows/pr-test.yml

@@ -946,60 +946,6 @@ jobs:
             fi
             python3 run_suite.py --suite quantization_test $RETRY_FLAG $CONTINUE_ON_ERROR_FLAG
 
-  unit-test-backend-1-gpu:
-    needs: [check-changes, call-gate, stage-a-test-1]
-    if: |
-      always() &&
-      (
-        (inputs.target_stage == 'unit-test-backend-1-gpu') ||
-        (
-          !inputs.target_stage &&
-          (github.event_name == 'schedule' || (!failure() && !cancelled())) &&
-          ((needs.check-changes.outputs.main_package == 'true') || (needs.check-changes.outputs.sgl_kernel == 'true'))
-        )
-      )
-    runs-on: 1-gpu-runner
-    env:
-      RUNNER_LABELS: 1-gpu-runner
-    strategy:
-      fail-fast: false
-      max-parallel: ${{ fromJson(needs.check-changes.outputs.max_parallel) }}
-      matrix:
-        part: [0, 1]
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-        with:
-          ref: ${{ inputs.pr_head_sha || inputs.ref || github.sha }}
-
-      - name: Download artifacts
-        if: needs.check-changes.outputs.sgl_kernel == 'true'
-        uses: actions/download-artifact@v4
-        with:
-          path: sgl-kernel/dist/
-          merge-multiple: true
-          pattern: wheel-python3.10-cuda12.9
-
-      - name: Install dependencies
-        timeout-minutes: 10
-        run: |
-          CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} bash scripts/ci/ci_install_dependency.sh
-
-      - name: Run test
-        timeout-minutes: 30
-        run: |
-          cd test/srt
-          RETRY_FLAG=""
-          if [[ "${{ needs.check-changes.outputs.enable_retry }}" == "true" ]]; then
-            RETRY_FLAG="--enable-retry"
-          fi
-          CONTINUE_ON_ERROR_FLAG=""
-          if [[ "${{ needs.check-changes.outputs.continue_on_error }}" == "true" ]]; then
-            CONTINUE_ON_ERROR_FLAG="--continue-on-error"
-          fi
-          python3 run_suite.py --suite per-commit-1-gpu --auto-partition-id ${{ matrix.part }} --auto-partition-size 2 $RETRY_FLAG $CONTINUE_ON_ERROR_FLAG
-
-
   stage-b-test-4-gpu-b200:
     needs: [check-changes, call-gate, stage-a-test-1, sgl-kernel-build-wheels]
     if: |
@@ -1047,9 +993,8 @@ jobs:
           fi
           IS_BLACKWELL=1 python3 run_suite.py --hw cuda --suite stage-b-test-4-gpu-b200 $CONTINUE_ON_ERROR_FLAG
 
-
   unit-test-backend-4-gpu:
-    needs: [check-changes, call-gate, unit-test-backend-1-gpu, stage-b-test-4-gpu-b200]
+    needs: [check-changes, call-gate, stage-b-test-small-1-gpu, stage-b-test-large-1-gpu, stage-b-test-large-2-gpu, stage-b-test-4-gpu-b200]
     if: |
       always() &&
       (
@@ -1101,7 +1046,7 @@ jobs:
           python3 run_suite.py --suite per-commit-4-gpu --auto-partition-id ${{ matrix.part }} --auto-partition-size 3 $RETRY_FLAG $CONTINUE_ON_ERROR_FLAG
 
   unit-test-backend-8-gpu-h200:
-    needs: [check-changes, call-gate, unit-test-backend-1-gpu, stage-b-test-4-gpu-b200]
+    needs: [check-changes, call-gate, stage-b-test-small-1-gpu, stage-b-test-large-1-gpu, stage-b-test-large-2-gpu, stage-b-test-4-gpu-b200]
     if: |
       always() &&
       (
@@ -1159,7 +1104,7 @@ jobs:
           python3 run_suite.py --suite per-commit-8-gpu-h200 --auto-partition-id ${{ matrix.part }} --auto-partition-size 4 $RETRY_FLAG $CONTINUE_ON_ERROR_FLAG
 
   unit-test-backend-8-gpu-h20:
-    needs: [check-changes, call-gate, unit-test-backend-1-gpu, stage-b-test-4-gpu-b200]
+    needs: [check-changes, call-gate, stage-b-test-small-1-gpu, stage-b-test-large-1-gpu, stage-b-test-large-2-gpu, stage-b-test-4-gpu-b200]
     if: |
       always() &&
       (
@@ -1406,7 +1351,7 @@ jobs:
           python3 -m unittest test_bench_serving.TestBenchServing.test_embeddings_api_batch_scaling
 
   performance-test-2-gpu:
-    needs: [check-changes, call-gate, unit-test-backend-1-gpu, stage-b-test-4-gpu-b200]
+    needs: [check-changes, call-gate, stage-b-test-4-gpu-b200]
     if: |
       always() &&
       (
@@ -1562,7 +1507,7 @@ jobs:
           python3 -m sglang.test.ci.run_with_retry test_moe_eval_accuracy_large.py
 
   unit-test-deepep-4-gpu:
-    needs: [check-changes, call-gate, unit-test-backend-1-gpu, stage-b-test-4-gpu-b200]
+    needs: [check-changes, call-gate, stage-b-test-small-1-gpu, stage-b-test-large-1-gpu, stage-b-test-large-2-gpu, stage-b-test-4-gpu-b200]
     if: |
       always() &&
       (
@@ -1610,7 +1555,7 @@ jobs:
           python3 run_suite.py --suite per-commit-4-gpu-deepep $RETRY_FLAG $CONTINUE_ON_ERROR_FLAG
 
   unit-test-deepep-8-gpu:
-    needs: [check-changes, call-gate, unit-test-backend-1-gpu, stage-b-test-4-gpu-b200]
+    needs: [check-changes, call-gate, stage-b-test-small-1-gpu, stage-b-test-large-1-gpu, stage-b-test-large-2-gpu, stage-b-test-4-gpu-b200]
     if: |
       always() &&
       (
@@ -1658,7 +1603,7 @@ jobs:
           python3 run_suite.py --suite per-commit-8-gpu-h200-deepep $RETRY_FLAG $CONTINUE_ON_ERROR_FLAG
 
   unit-test-backend-4-gpu-b200:
-    needs: [check-changes, call-gate, unit-test-backend-1-gpu, stage-b-test-4-gpu-b200]
+    needs: [check-changes, call-gate, stage-b-test-small-1-gpu, stage-b-test-large-1-gpu, stage-b-test-large-2-gpu, stage-b-test-4-gpu-b200]
     if: |
       always() &&
       (
@@ -1711,7 +1656,7 @@ jobs:
           IS_BLACKWELL=1 python3 run_suite.py --suite per-commit-4-gpu-b200 --auto-partition-id ${{ matrix.part }} --auto-partition-size 3 --timeout-per-file 1800 $RETRY_FLAG $CONTINUE_ON_ERROR_FLAG
 
   unit-test-backend-4-gpu-gb200:
-    needs: [check-changes, call-gate, unit-test-backend-1-gpu, stage-b-test-4-gpu-b200, sgl-kernel-build-wheels-arm]
+    needs: [check-changes, call-gate, stage-b-test-small-1-gpu, stage-b-test-large-1-gpu, stage-b-test-large-2-gpu, stage-b-test-4-gpu-b200, sgl-kernel-build-wheels-arm]
     if: |
       always() &&
       (
@@ -1784,7 +1729,6 @@ jobs:
         stage-b-test-large-2-gpu,
         stage-c-test-large-4-gpu,
         quantization-test,
-        unit-test-backend-1-gpu,
         stage-b-test-4-gpu-b200,
         unit-test-backend-4-gpu,
         unit-test-backend-8-gpu-h20,

scripts/ci/slash_command_handler.py

@@ -154,7 +154,6 @@ def handle_rerun_stage(
         "multimodal-gen-test-1-gpu",
         "multimodal-gen-test-2-gpu",
         "quantization-test",
-        "unit-test-backend-1-gpu",
         "stage-b-test-4-gpu-b200",
         "unit-test-backend-4-gpu",
         "unit-test-backend-8-gpu-h200",

test/registered/vlm/test_evs.py

@@ -3,8 +3,11 @@ from types import SimpleNamespace
 
 import pytest
 
+from sglang.test.ci.ci_register import register_cuda_ci
 from sglang.test.test_utils import run_doctests
 
+register_cuda_ci(est_time=20, suite="stage-b-test-small-1-gpu")
+
 
 def test_resolve_evs_config():
     from sglang.srt.multimodal.evs import EVS, EVSConfig, EVSProcessor

test/registered/vlm/test_video_utils.py

@@ -3,6 +3,9 @@ from dataclasses import dataclass
 import pytest
 
 from sglang.srt.utils import sample_video_frames
+from sglang.test.ci.ci_register import register_cpu_ci
+
+register_cpu_ci(est_time=5, suite="stage-a-cpu-only")
 
 
 class DummyVideo:

test/srt/run_suite.py

@@ -8,10 +8,6 @@ from sglang.test.ci.ci_utils import TestFile, run_unittest_files
 
 # NOTE: please sort the test cases alphabetically by the test file name
 suites = {
-    "per-commit-1-gpu": [
-        TestFile("test_evs.py", 20),
-        TestFile("test_video_utils.py", 5),
-    ],
     "per-commit-4-gpu": [
         TestFile("models/test_qwen3_next_models.py", 650),
         TestFile("test_gpt_oss_4gpu.py", 300),
@@ -91,7 +87,6 @@ suite_amd = {
         TestFile("test_rope_rocm.py", 3),
         # TestFile("test_torch_compile_moe.py", 210), # Disabled temporarily, see https://github.com/sgl-project/sglang/issues/13107
         TestFile("test_type_based_dispatcher.py", 10),
-        TestFile("test_video_utils.py", 8),
         # Disabled temporarily
         # TestFile("test_vlm_input_format.py", 300),
         # TestFile("openai_server/features/test_openai_server_hidden_states.py", 240),