diff --git a/examples/auto_deploy/nano_v3.yaml b/examples/auto_deploy/nano_v3.yaml
index a87e26242546..a847d9a8d4ed 100644
--- a/examples/auto_deploy/nano_v3.yaml
+++ b/examples/auto_deploy/nano_v3.yaml
@@ -14,6 +14,7 @@ kv_cache_config:
 transforms:
   detect_sharding:
     sharding_dims: ['ep', 'bmm']
+    allreduce_strategy: 'AUTO'
     manual_config:
       head_dim: 128
       tp_plan:
diff --git a/tensorrt_llm/_torch/auto_deploy/config/default.yaml b/tensorrt_llm/_torch/auto_deploy/config/default.yaml
index a7251de20a12..4edf3de15072 100644
--- a/tensorrt_llm/_torch/auto_deploy/config/default.yaml
+++ b/tensorrt_llm/_torch/auto_deploy/config/default.yaml
@@ -81,7 +81,7 @@ transforms:
     sharding_source: ['manual', 'factory', 'heuristic']
     support_partial_config: true
     sharding_dims: ['tp', 'ep', 'bmm']
-    allreduce_strategy: 'AUTO'
+    allreduce_strategy: 'NCCL'
     dist_backend: auto
     requires_shape_prop: true
   sharding_transform_executor: