chore: bump sgl-kernel version to 0.3.17.post2 (#13542)

.github/workflows/pr-test.yml

@@ -88,6 +88,8 @@ jobs:
         run: |
           cd sgl-kernel
           ./build.sh "${{ matrix.python-version }}" "${{ matrix.cuda-version }}"
+        env:
+          USE_CCACHE: 1
 
       - name: Upload artifacts
         uses: actions/upload-artifact@v4
@@ -127,6 +129,8 @@ jobs:
         run: |
           cd sgl-kernel
           ./build.sh "${{ matrix.python-version }}" "${{ matrix.cuda-version }}"
+        env:
+          USE_CCACHE: 1
 
       - name: Upload artifacts
         uses: actions/upload-artifact@v4
@@ -241,23 +245,6 @@ jobs:
           echo "All benchmark tests completed!"
 
   # =============================================== multimodal_gen ====================================================
-  multimodal-gen-test:
-    needs: [check-changes]
-    if: needs.check-changes.outputs.multimodal_gen == 'true'
-    runs-on: 1-gpu-runner
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-
-      - 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_performance.py
 
   # Adding a single CUDA13 smoke test to verify that the kernel builds and runs
   # TODO: Add back this test when it can pass on CI
@@ -328,6 +315,33 @@ jobs:
         run: |
           rm -rf python/sglang/logs || true
 
+
+  multimodal-gen-test:
+    needs: [check-changes, stage-a-test-1, sgl-kernel-build-wheels]
+    if: needs.check-changes.outputs.multimodal_gen == 'true'
+    runs-on: 1-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_performance.py
+
   unit-test-backend-1-gpu:
     needs: [check-changes, stage-a-test-1, sgl-kernel-build-wheels]
     if: always() && !failure() && !cancelled() &&

scripts/ci/validate_and_download_models.py

@@ -39,6 +39,7 @@ except ImportError:
 RUNNER_LABEL_MODEL_MAP: Dict[str, List[str]] = {
     "1-gpu-runner": [
         "Alibaba-NLP/gte-Qwen2-1.5B-instruct",
+        "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct",
         "deepseek-ai/DeepSeek-OCR",
         "google/gemma-3-4b-it",
         "intfloat/e5-mistral-7b-instruct",

sgl-kernel/pyproject.toml

@@ -8,7 +8,7 @@ build-backend = "scikit_build_core.build"
 
 [project]
 name = "sgl-kernel"
-version = "0.3.17.post1"
+version = "0.3.17.post2"
 description = "Kernel Library for SGLang"
 readme = "README.md"
 requires-python = ">=3.10"

sgl-kernel/pyproject_cpu.toml

@@ -8,7 +8,7 @@ build-backend = "scikit_build_core.build"
 
 [project]
 name = "sgl-kernel"
-version = "0.3.17.post1"
+version = "0.3.17.post2"
 description = "Kernel Library for SGLang"
 readme = "README.md"
 requires-python = ">=3.10"

sgl-kernel/pyproject_rocm.toml

@@ -9,7 +9,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "sgl-kernel"
-version = "0.3.17.post1"
+version = "0.3.17.post2"
 description = "Kernel Library for SGLang"
 readme = "README.md"
 requires-python = ">=3.10"

sgl-kernel/python/sgl_kernel/version.py

@@ -1 +1 @@
-__version__ = "0.3.17.post1"
+__version__ = "0.3.17.post2"