diff --git a/CHANGELOG.md b/CHANGELOG.md
index 92aca1e0774..3f30ab1d83e 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -53,6 +53,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Pearson metrics now only store 6 statistics instead of all predictions and targets ([#380](https://github.com/PyTorchLightning/metrics/pull/380))
 
 
+- Use `torch.argmax` instead of `torch.topk` when `k=1` for better performance ([#419](https://github.com/PyTorchLightning/metrics/pull/419))
+
+
 ### Deprecated
 
 - Rename `r2score` >> `r2_score` and `kldivergence` >> `kl_divergence` in `functional` ([#371](https://github.com/PyTorchLightning/metrics/pull/371))
diff --git a/torchmetrics/utilities/data.py b/torchmetrics/utilities/data.py
index 40f81f77ccb..46648352e8f 100644
--- a/torchmetrics/utilities/data.py
+++ b/torchmetrics/utilities/data.py
@@ -94,7 +94,10 @@ def select_topk(prob_tensor: Tensor, topk: int = 1, dim: int = 1) -> Tensor:
                 [1, 1, 0]], dtype=torch.int32)
     """
     zeros = torch.zeros_like(prob_tensor)
-    topk_tensor = zeros.scatter(dim, prob_tensor.topk(k=topk, dim=dim).indices, 1.0)
+    if topk == 1:  # argmax has better performance than topk
+        topk_tensor = zeros.scatter(dim, prob_tensor.argmax(dim=dim, keepdim=True), 1.0)
+    else:
+        topk_tensor = zeros.scatter(dim, prob_tensor.topk(k=topk, dim=dim).indices, 1.0)
     return topk_tensor.int()