[AMD] Add GLM-4.7-FP8 accuracy CI test for MI35x

.github/workflows/nightly-test-amd-rocm720.yml

@@ -61,6 +61,7 @@ on:
           - nightly-8-gpu-mi35x-qwen3-235b-mxfp4-rocm720
           - nightly-8-gpu-mi35x-qwen35-rocm720
           - nightly-8-gpu-mi35x-glm5-rocm720
+          - nightly-8-gpu-mi35x-glm47-fp8-rocm720
           - nightly-8-gpu-mi35x-minimax-m25-rocm720
       job_filter:
         description: 'Or type comma-separated job names (overrides dropdown if non-empty)'
@@ -1272,6 +1273,39 @@ jobs:
           echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
           exit ${TEST_EXIT_CODE:-0}
 
+  # MI35x 8-GPU GLM-4.7-FP8 (Accuracy) ROCm 7.2
+  nightly-8-gpu-mi35x-glm47-fp8-rocm720:
+    if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') && (!(inputs.job_filter || inputs.job_select) || (inputs.job_filter || inputs.job_select) == 'all' || contains(format(',{0},', inputs.job_filter || inputs.job_select), ',nightly-8-gpu-mi35x-glm47-fp8-rocm720,'))
+    runs-on: linux-mi35x-gpu-8
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ inputs.ref || github.ref }}
+
+      - name: Setup docker (ROCm 7.2)
+        run: |
+          touch github_summary.md
+          bash scripts/ci/amd/amd_ci_start_container.sh --rocm-version rocm720
+        env:
+          GITHUB_WORKSPACE: ${{ github.workspace }}
+
+      - name: Install dependencies
+        run: |
+          bash scripts/ci/amd/amd_ci_install_dependency.sh
+          # Install tabulate for run_suite.py (missing in MI35x container)
+          bash scripts/ci/amd/amd_ci_exec.sh pip install tabulate
+
+      - name: Accuracy Test MI35x ROCm 7.2 (8-GPU GLM-4.7-FP8)
+        timeout-minutes: 120
+        run: |
+          > github_summary.md  # Clear summary file
+          bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
+            -e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
+            python3 run_suite.py --hw amd --suite nightly-amd-8-gpu-mi35x-glm47-fp8 --nightly --timeout-per-file 3600 ${{ inputs.continue_on_error && '--continue-on-error' || '' }} || TEST_EXIT_CODE=$?
+          echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
+          exit ${TEST_EXIT_CODE:-0}
+
   # MI35x 8-GPU MiniMax-M2.5 (Accuracy) ROCm 7.2
   nightly-8-gpu-mi35x-minimax-m25-rocm720:
     if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') && (!(inputs.job_filter || inputs.job_select) || (inputs.job_filter || inputs.job_select) == 'all' || contains(format(',{0},', inputs.job_filter || inputs.job_select), ',nightly-8-gpu-mi35x-minimax-m25-rocm720,'))
@@ -1382,6 +1416,7 @@ jobs:
       - nightly-8-gpu-mi35x-qwen3-235b-mxfp4-rocm720
       - nightly-8-gpu-mi35x-qwen35-rocm720
       - nightly-8-gpu-mi35x-glm5-rocm720
+      - nightly-8-gpu-mi35x-glm47-fp8-rocm720
       - nightly-8-gpu-mi35x-minimax-m25-rocm720
     runs-on: ubuntu-latest
     steps:

test/registered/amd/accuracy/mi35x/test_glm47_fp8_eval_mi35x.py

@@ -0,0 +1,61 @@
+"""MI35x GLM-4.7-FP8 GSM8K Accuracy Evaluation Test (8-GPU)
+
+Tests GLM-4.7-FP8 accuracy using GSM8K benchmark on MI35x.
+
+Registry: nightly-amd-8-gpu-mi35x-glm47-fp8 suite
+"""
+
+import os
+
+# Set HF cache for MI35x
+os.environ.setdefault("HF_HOME", "/data2/models/huggingface")
+os.environ.setdefault("HF_HUB_CACHE", "/data2/models/huggingface/hub")
+
+import unittest
+
+from sglang.test.accuracy_test_runner import AccuracyTestParams
+from sglang.test.ci.ci_register import register_amd_ci
+from sglang.test.run_combined_tests import run_combined_tests
+from sglang.test.test_utils import ModelLaunchSettings
+
+# Register for AMD CI - MI35x GLM-4.7-FP8 accuracy test (~30 min)
+register_amd_ci(
+    est_time=1800,
+    suite="nightly-amd-8-gpu-mi35x-glm47-fp8",
+    nightly=True,
+)
+
+GLM_4_7_FP8_MODEL_PATH = "zai-org/GLM-4.7-FP8"
+
+
+class TestGLM47FP8EvalMI35x(unittest.TestCase):
+    """GLM-4.7-FP8 GSM8K Accuracy Evaluation Test for MI35x."""
+
+    def test_glm_47_fp8(self):
+        """Run accuracy test for GLM-4.7-FP8."""
+        base_args = [
+            "--trust-remote-code",
+            "--tool-call-parser=glm47",
+            "--reasoning-parser=glm45",
+        ]
+
+        variants = [
+            ModelLaunchSettings(
+                GLM_4_7_FP8_MODEL_PATH,
+                tp_size=8,
+                extra_args=base_args,
+                variant="TP8",
+            ),
+        ]
+
+        run_combined_tests(
+            models=variants,
+            test_name="GLM-4.7-FP8",
+            accuracy_params=AccuracyTestParams(
+                dataset="gsm8k", baseline_accuracy=0.92
+            ),
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()