diff --git a/src/strands/models/ollama.py b/src/strands/models/ollama.py
index 54805ac16..cf7108c3a 100644
--- a/src/strands/models/ollama.py
+++ b/src/strands/models/ollama.py
@@ -280,7 +280,7 @@ def format_chunk(self, event: dict[str, Any]) -> StreamEvent:
                             "totalTokens": event["data"].eval_count + event["data"].prompt_eval_count,
                         },
                         "metrics": {
-                            "latencyMs": event["data"].total_duration / 1e6,
+                            "latencyMs": int(event["data"].total_duration / 1e6),
                         },
                     },
                 }
diff --git a/tests/strands/models/test_ollama.py b/tests/strands/models/test_ollama.py
index 7a6bbf97c..360683d08 100644
--- a/tests/strands/models/test_ollama.py
+++ b/tests/strands/models/test_ollama.py
@@ -407,7 +407,7 @@ def test_format_chunk_metadata(model):
                 "totalTokens": 150,
             },
             "metrics": {
-                "latencyMs": 1.0,
+                "latencyMs": 1,
             },
         },
     }
@@ -447,7 +447,7 @@ async def test_stream(ollama_client, model, agenerator, alist, captured_warnings
         {
             "metadata": {
                 "usage": {"inputTokens": 5, "outputTokens": 10, "totalTokens": 15},
-                "metrics": {"latencyMs": 1.0},
+                "metrics": {"latencyMs": 1},
             }
         },
     ]
@@ -525,7 +525,7 @@ async def test_stream_with_tool_calls(ollama_client, model, agenerator, alist):
     assert tru_events[8] == {
         "metadata": {
             "usage": {"inputTokens": 8, "outputTokens": 15, "totalTokens": 23},
-            "metrics": {"latencyMs": 2.0},
+            "metrics": {"latencyMs": 2},
         }
     }
     expected_request = {