diff --git a/.buildkite/amd/test-amd-ready.yml b/.buildkite/amd/test-amd-ready.yml
index 695f47dc1..585f089a2 100644
--- a/.buildkite/amd/test-amd-ready.yml
+++ b/.buildkite/amd/test-amd-ready.yml
@@ -107,6 +107,11 @@ steps:
         - export VLLM_LOGGING_LEVEL=DEBUG
         - export VLLM_WORKER_MULTIPROC_METHOD=spawn
         - timeout 20m pytest -s -v tests/diffusion/distributed/test_sequence_parallel.py -m core_model
+        - >-
+          timeout 20m pytest -s -v
+          tests/diffusion/models/ltx2/test_ltx2_transformer_ulysses.py
+          -m 'core_model and rocm and cards_2'
+          --run-level "core_model"
 
     - label: "Diffusion GPU Worker Test"
       agent_pool: mi300_2
@@ -122,7 +127,13 @@ steps:
       mirror_hardwares: [amdproduction]
       grade: Blocking
       commands:
-        - timeout 15m pytest -sv tests/diffusion/models/ -m 'core_model and cuda' --run-level "core_model"
+        # Preserve legacy one-GPU tests without cards_1 while excluding every
+        # registered multi-GPU marker from this one-GPU job.
+        - >-
+          timeout 15m pytest -sv tests/diffusion/models/
+          -m 'core_model and cuda and not
+          (cards_2 or cards_3 or cards_4 or cards_5 or cards_6 or cards_7 or cards_8)'
+          --run-level "core_model"
 
     - label: "Diffusion · Wan22 Test"
       agent_pool: mi300_1
diff --git a/tests/diffusion/models/ltx2/test_ltx2_transformer_ulysses.py b/tests/diffusion/models/ltx2/test_ltx2_transformer_ulysses.py
index 9f09b72a7..0e080e015 100644
--- a/tests/diffusion/models/ltx2/test_ltx2_transformer_ulysses.py
+++ b/tests/diffusion/models/ltx2/test_ltx2_transformer_ulysses.py
@@ -156,10 +156,14 @@ def _run_transformer_parity(rank: int, world_size: int, master_port: int) -> Non
 
 
 @pytest.mark.full_model
-@hardware_test(res={"cuda": "L4"}, num_cards=2)
+@hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards=2)
 def test_ltx_ulysses_transformer_matches_sp1(unused_tcp_port) -> None:
     """Exercise real plan hooks, RoPE/timestep sharding, and attention routing."""
     world_size = 2
+    device_count = current_omni_platform.device_count()
+    assert device_count >= world_size, (
+        f"LTX Ulysses parity requires {world_size} accelerator devices; only {device_count} are visible"
+    )
     torch.multiprocessing.spawn(
         _run_transformer_parity,
         args=(world_size, unused_tcp_port),