[diffusion] fix: fix hunyuanvideo and add 2-gpu ci test (#13720)

Co-authored-by: Mick <mickjagger19@icloud.com>

.github/workflows/pr-test.yml

@@ -351,10 +351,15 @@ jobs:
           python3 run_suite.py --hw cpu --suite default
 
 
-  multimodal-gen-test:
+  multimodal-gen-test-1-gpu:
     needs: [check-changes, sgl-kernel-build-wheels]
     if: (always() && !failure() && !cancelled()) && needs.check-changes.outputs.multimodal_gen == 'true'
     runs-on: 1-gpu-runner
+    strategy:
+      fail-fast: false
+      max-parallel: 5
+      matrix:
+        test_file: ["test_server_perf_a.py", "test_server_perf_b.py"]
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
@@ -375,7 +380,34 @@ jobs:
         timeout-minutes: 60
         run: |
           cd python
-          pytest -s -v --log-cli-level=INFO sglang/multimodal_gen/test/server/test_server_performance.py
+          pytest -s -v --log-cli-level=INFO sglang/multimodal_gen/test/server/${{ matrix.test_file }}
+
+
+  multimodal-gen-test-2-gpu:
+    needs: [check-changes, sgl-kernel-build-wheels]
+    if: (always() && !failure() && !cancelled()) && needs.check-changes.outputs.multimodal_gen == 'true'
+    runs-on: 2-gpu-runner
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Download artifacts
+        if: needs.check-changes.outputs.sgl_kernel == 'true'
+        uses: actions/download-artifact@v4
+        with:
+          path: sgl-kernel/dist/
+          merge-multiple: true
+          pattern: wheel-python3.10-cuda12.9
+
+      - name: Install dependencies
+        run: |
+          CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} bash scripts/ci/ci_install_dependency.sh diffusion
+
+      - name: Run diffusion server tests
+        timeout-minutes: 60
+        run: |
+          cd python
+          pytest -s -v --log-cli-level=INFO sglang/multimodal_gen/test/server/test_server_perf_2_gpu.py
 
   unit-test-backend-1-gpu:
     needs: [check-changes, stage-a-test-1]
@@ -963,7 +995,8 @@ jobs:
         sgl-kernel-benchmark-test,
         sgl-kernel-b200-test,
 
-        multimodal-gen-test,
+        multimodal-gen-test-1-gpu,
+        multimodal-gen-test-2-gpu,
 
         stage-a-test-1,
         unit-test-backend-1-gpu,