diff --git a/test/registered/moe/test_zero_experts.py b/test/registered/moe/test_zero_experts.py
index 0754baa59..8b00448e4 100644
--- a/test/registered/moe/test_zero_experts.py
+++ b/test/registered/moe/test_zero_experts.py
@@ -3,10 +3,11 @@ import unittest
 import torch
 
 from sglang.kernels.ops.moe.ep_moe_kernels import zero_experts_compute_triton
-from sglang.test.ci.ci_register import register_cuda_ci
+from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cuda_ci(est_time=5, stage="base-b", runner_config="1-gpu-small")
+register_amd_ci(est_time=5, stage="stage-b", runner_config="1-gpu-small-amd")
 
 
 @unittest.skipIf(not torch.cuda.is_available(), "CUDA is required")
diff --git a/test/registered/unit/distributed/test_cuda_wrapper.py b/test/registered/unit/distributed/test_cuda_wrapper.py
index e7383c63e..d068bffe6 100644
--- a/test/registered/unit/distributed/test_cuda_wrapper.py
+++ b/test/registered/unit/distributed/test_cuda_wrapper.py
@@ -2,9 +2,10 @@ import io
 
 import pytest
 
-from sglang.test.ci.ci_register import register_cuda_ci
+from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
 
 register_cuda_ci(est_time=2, stage="base-b", runner_config="1-gpu-small")
+register_amd_ci(est_time=2, stage="stage-b", runner_config="1-gpu-small-amd")
 
 from sglang.srt.distributed.device_communicators import cuda_wrapper