diff --git a/pytorch_lightning/plugins/precision/sharded_native_amp.py b/pytorch_lightning/plugins/precision/sharded_native_amp.py
index 12ae5d0bc6be3..39dc01f97df11 100644
--- a/pytorch_lightning/plugins/precision/sharded_native_amp.py
+++ b/pytorch_lightning/plugins/precision/sharded_native_amp.py
@@ -33,5 +33,8 @@ def __init__(self) -> None:
         self.scaler = ShardedGradScaler()
 
     def clip_gradients(self, optimizer: 'Optimizer', clip_val: Union[int, float], norm_type: float = 2.0) -> None:
+        if clip_val <= 0:
+            return
+
         optimizer = cast(OSS, optimizer)
         optimizer.clip_grad_norm(clip_val, norm_type=norm_type)
diff --git a/tests/plugins/test_sharded_plugin.py b/tests/plugins/test_sharded_plugin.py
index 9623aa4d0265c..b59563f70e4aa 100644
--- a/tests/plugins/test_sharded_plugin.py
+++ b/tests/plugins/test_sharded_plugin.py
@@ -1,4 +1,5 @@
 import os
+from unittest import mock
 
 import pytest
 import torch
@@ -11,6 +12,22 @@
 from tests.helpers.runif import RunIf
 
 
+@pytest.mark.parametrize("clip_val", [0, 10])
+@RunIf(min_gpus=1, skip_windows=True, amp_native=True, fairscale=True)
+@mock.patch('fairscale.optim.oss.OSS.clip_grad_norm')
+def test_ddp_sharded_precision_16_clip_gradients(mock_oss_clip_grad_norm, clip_val, tmpdir):
+    """
+    Ensure that clip gradients is only called if the value is greater than 0.
+    """
+    model = BoringModel()
+    trainer = Trainer(accelerator='ddp_sharded', gpus=1, precision=16, fast_dev_run=True, gradient_clip_val=clip_val)
+    trainer.fit(model)
+    if clip_val > 0:
+        mock_oss_clip_grad_norm.assert_called()
+    else:
+        mock_oss_clip_grad_norm.assert_not_called()
+
+
 @RunIf(fairscale=True)
 @pytest.mark.parametrize(["accelerator"], [("ddp_sharded", ), ("ddp_sharded_spawn", )])
 def test_sharded_ddp_choice(tmpdir, accelerator):