diff --git a/py/torch_tensorrt/dynamo/conversion/custom_ops_converters.py b/py/torch_tensorrt/dynamo/conversion/custom_ops_converters.py
index 7bd70705a0..36386b8d7b 100644
--- a/py/torch_tensorrt/dynamo/conversion/custom_ops_converters.py
+++ b/py/torch_tensorrt/dynamo/conversion/custom_ops_converters.py
@@ -28,7 +28,9 @@
     _LOGGER.info("Using native TensorRT DistCollective API for distributed operations")
 
     @dynamo_tensorrt_converter(
-        tensorrt_fused_nccl_all_gather_op, requires_native_multidevice=True
+        tensorrt_fused_nccl_all_gather_op,
+        requires_native_multidevice=True,
+        supports_dynamic_shapes=True,
     )
     def fused_nccl_all_gather(
         ctx: ConversionContext,
@@ -47,7 +49,9 @@ def fused_nccl_all_gather(
         )
 
     @dynamo_tensorrt_converter(
-        tensorrt_fused_nccl_reduce_scatter_op, requires_native_multidevice=True
+        tensorrt_fused_nccl_reduce_scatter_op,
+        requires_native_multidevice=True,
+        supports_dynamic_shapes=True,
     )
     def fused_nccl_reduce_scatter(
         ctx: ConversionContext,
@@ -69,7 +73,9 @@ def fused_nccl_reduce_scatter(
         )
 
     @dynamo_tensorrt_converter(
-        tensorrt_fused_nccl_all_reduce_op, requires_native_multidevice=True
+        tensorrt_fused_nccl_all_reduce_op,
+        requires_native_multidevice=True,
+        supports_dynamic_shapes=True,
     )
     def fused_nccl_all_reduce(
         ctx: ConversionContext,
@@ -90,7 +96,9 @@ def fused_nccl_all_reduce(
         )
 
     @dynamo_tensorrt_converter(
-        tensorrt_fused_nccl_all_to_all_op, requires_native_multidevice=True
+        tensorrt_fused_nccl_all_to_all_op,
+        requires_native_multidevice=True,
+        supports_dynamic_shapes=True,
     )
     def fused_nccl_all_to_all(
         ctx: ConversionContext,
@@ -109,7 +117,9 @@ def fused_nccl_all_to_all(
         )
 
     @dynamo_tensorrt_converter(
-        tensorrt_fused_nccl_scatter_op, requires_native_multidevice=True
+        tensorrt_fused_nccl_scatter_op,
+        requires_native_multidevice=True,
+        supports_dynamic_shapes=True,
     )
     def fused_nccl_scatter(
         ctx: ConversionContext,
@@ -125,7 +135,9 @@ def fused_nccl_scatter(
         )
 
     @dynamo_tensorrt_converter(
-        tensorrt_fused_nccl_gather_op, requires_native_multidevice=True
+        tensorrt_fused_nccl_gather_op,
+        requires_native_multidevice=True,
+        supports_dynamic_shapes=True,
     )
     def fused_nccl_gather(
         ctx: ConversionContext,