diff --git a/deepmd/pt/model/atomic_model/pairtab_atomic_model.py b/deepmd/pt/model/atomic_model/pairtab_atomic_model.py
index 6933fdc19a..a77e5391f8 100644
--- a/deepmd/pt/model/atomic_model/pairtab_atomic_model.py
+++ b/deepmd/pt/model/atomic_model/pairtab_atomic_model.py
@@ -392,6 +392,11 @@ def _get_pairwise_dist(coords: torch.Tensor, nlist: torch.Tensor) -> torch.Tenso
         -------
         torch.Tensor
             The pairwise distance between the atoms (nframes, nloc, nnei).
+
+        Notes
+        -----
+        Safe gradient implementation: when diff is zero (padding entries),
+            both distance and gradient are zero.
         """
         nframes, nloc, nnei = nlist.shape
         coord_l = coords[:, :nloc].view(nframes, -1, 1, 3)
@@ -399,7 +404,17 @@ def _get_pairwise_dist(coords: torch.Tensor, nlist: torch.Tensor) -> torch.Tenso
         coord_r = torch.gather(coords, 1, index)
         coord_r = coord_r.view(nframes, nloc, nnei, 3)
         diff = coord_r - coord_l
-        pairwise_rr = torch.linalg.norm(diff, dim=-1, keepdim=True).squeeze(-1)
+        diff_sq = torch.sum(diff * diff, dim=-1, keepdim=True)
+
+        # When diff is zero, output is zero and gradient is also zero
+        mask = diff_sq.squeeze(-1) > 0
+        pairwise_rr = torch.where(
+            mask.unsqueeze(-1),
+            torch.sqrt(
+                torch.where(mask.unsqueeze(-1), diff_sq, torch.ones_like(diff_sq))
+            ),
+            torch.zeros_like(diff_sq),
+        ).squeeze(-1)
         return pairwise_rr
 
     @staticmethod