From e990cd01a9123b55068c81069dcc552609a0ea48 Mon Sep 17 00:00:00 2001
From: Andrei Rusu <rusu.c.andrei@gmail.com>
Date: Tue, 24 Feb 2026 13:38:51 +0200
Subject: [PATCH 1/2] fix(evals): exact match float int mismatch

---
 .../eval/evaluators/exact_match_evaluator.py  | 11 ++--
 tests/evaluators/test_evaluator_methods.py    | 61 +++++++++++++++++++
 2 files changed, 68 insertions(+), 4 deletions(-)

diff --git a/src/uipath/eval/evaluators/exact_match_evaluator.py b/src/uipath/eval/evaluators/exact_match_evaluator.py
index 5cad0670b..cd422eb23 100644
--- a/src/uipath/eval/evaluators/exact_match_evaluator.py
+++ b/src/uipath/eval/evaluators/exact_match_evaluator.py
@@ -58,11 +58,14 @@ async def evaluate(
         """
         actual_output = str(self._get_actual_output(agent_execution))
         expected_output = str(self._get_expected_output(evaluation_criteria))
-        if not self.evaluator_config.case_sensitive:
-            actual_output = actual_output.lower()
-            expected_output = expected_output.lower()
 
-        is_exact_match = actual_output == expected_output
+        try:
+            is_exact_match = float(actual_output) == float(expected_output)
+        except ValueError:
+            if not self.evaluator_config.case_sensitive:
+                actual_output = actual_output.lower()
+                expected_output = expected_output.lower()
+            is_exact_match = actual_output == expected_output
         if self.evaluator_config.negated:
             is_exact_match = not is_exact_match
 
diff --git a/tests/evaluators/test_evaluator_methods.py b/tests/evaluators/test_evaluator_methods.py
index 7b09fb8d7..3aaa99368 100644
--- a/tests/evaluators/test_evaluator_methods.py
+++ b/tests/evaluators/test_evaluator_methods.py
@@ -199,6 +199,67 @@ async def test_exact_match_negated(
         assert isinstance(result, NumericEvaluationResult)
         assert result.score == 0.0
 
+    @pytest.mark.asyncio
+    @pytest.mark.parametrize(
+        "actual, expected",
+        [
+            ("1.0", "1"),
+            ("1", "1.0"),
+            ("1e0", "1"),
+            ("1.00", "1.0"),
+            ("0.5", "0.50"),
+            ("-3.0", "-3"),
+        ],
+    )
+    async def test_exact_match_numeric_leniency(
+        self, actual: str, expected: str
+    ) -> None:
+        """Test that numerically equal values match regardless of string representation."""
+        execution = AgentExecution(
+            agent_input={"input": "Test"},
+            agent_output={"result": actual},
+            agent_trace=[],
+        )
+        config = {
+            "name": "ExactMatchNumericTest",
+            "case_sensitive": True,
+            "target_output_key": "result",
+        }
+        evaluator = ExactMatchEvaluator.model_validate(
+            {"evaluatorConfig": config, "id": str(uuid.uuid4())}
+        )
+        criteria = OutputEvaluationCriteria(expected_output={"result": expected})  # pyright: ignore[reportCallIssue]
+
+        result = await evaluator.evaluate(execution, criteria)
+
+        assert isinstance(result, NumericEvaluationResult)
+        assert result.score == 1.0, (
+            f"Expected '{actual}' and '{expected}' to be considered equal as numbers"
+        )
+
+    @pytest.mark.asyncio
+    async def test_exact_match_numeric_non_equal(self) -> None:
+        """Test that numerically different values do not match."""
+        execution = AgentExecution(
+            agent_input={"input": "Test"},
+            agent_output={"result": "1.5"},
+            agent_trace=[],
+        )
+        config = {
+            "name": "ExactMatchNumericTest",
+            "case_sensitive": True,
+            "target_output_key": "result",
+        }
+        evaluator = ExactMatchEvaluator.model_validate(
+            {"evaluatorConfig": config, "id": str(uuid.uuid4())}
+        )
+        criteria = OutputEvaluationCriteria(expected_output={"result": "1"})  # pyright: ignore[reportCallIssue]
+
+        result = await evaluator.evaluate(execution, criteria)
+
+        assert isinstance(result, NumericEvaluationResult)
+        assert result.score == 0.0
+
     @pytest.mark.asyncio
     async def test_exact_match_validate_and_evaluate_criteria(
         self, sample_agent_execution: AgentExecution

From 73395d31c5b0dec4cb5d21e5f0d7a7fe58940bd3 Mon Sep 17 00:00:00 2001
From: Andrei Rusu <rusu.c.andrei@gmail.com>
Date: Tue, 24 Feb 2026 15:46:59 +0200
Subject: [PATCH 2/2] fix(evals): bump version

---
 pyproject.toml | 2 +-
 uv.lock        | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index c796a2073..a9032f2ba 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "uipath"
-version = "2.9.3"
+version = "2.9.4"
 description = "Python SDK and CLI for UiPath Platform, enabling programmatic interaction with automation services, process management, and deployment tools."
 readme = { file = "README.md", content-type = "text/markdown" }
 requires-python = ">=3.11"
diff --git a/uv.lock b/uv.lock
index e108fcc76..ecd62cd50 100644
--- a/uv.lock
+++ b/uv.lock
@@ -2531,7 +2531,7 @@ wheels = [
 
 [[package]]
 name = "uipath"
-version = "2.9.3"
+version = "2.9.4"
 source = { editable = "." }
 dependencies = [
     { name = "applicationinsights" },