[ROCm][CI] Enable hybrid chunked prefill test (#38317)

Signed-off-by: Andreas Karatzas <akaratza@amd.com>

.buildkite/test-amd.yaml

@@ -1803,6 +1803,19 @@ steps:
     - pytest -v -s v1/e2e/spec_decode/test_spec_decode.py -k "eagle_correctness_heavy"
   
 
+- label: V1 e2e (4xH100-4xMI325) # TBD
+  timeout_in_minutes: 180
+  mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi325]
+  agent_pool: mi325_4
+  optional: true
+  source_file_dependencies:
+    - vllm/v1/attention/backends/utils.py
+    - vllm/v1/worker/gpu_model_runner.py
+    - tests/v1/e2e/test_hybrid_chunked_prefill.py
+  commands:
+    - pytest -v -s v1/e2e/test_hybrid_chunked_prefill.py
+
+
 - label: V1 Spec Decode # TBD
   timeout_in_minutes: 40
   mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi325]

tests/v1/e2e/test_hybrid_chunked_prefill.py

@@ -36,14 +36,20 @@ MESSAGES = [
 ]
 
 
-@pytest.mark.skipif(not current_platform.is_cuda(), reason="CUDA not available")
 @pytest.mark.parametrize(
     "model_name",
     [
         pytest.param("Qwen/Qwen3.5-4B", marks=[large_gpu_mark(min_gb=40)]),
         pytest.param(
             "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8",
-            marks=[large_gpu_mark(min_gb=80)] + multi_gpu_marks(num_gpus=4),
+            marks=[large_gpu_mark(min_gb=80)]
+            + multi_gpu_marks(num_gpus=4)
+            + [
+                pytest.mark.skipif(
+                    not current_platform.is_cuda(),
+                    reason="modelopt quantization is supported only on CUDA",
+                )
+            ],
         ),
     ],
 )