diff --git a/crates/vm/src/builtins/float.rs b/crates/vm/src/builtins/float.rs
index ddce34a88..5a64a6e64 100644
--- a/crates/vm/src/builtins/float.rs
+++ b/crates/vm/src/builtins/float.rs
@@ -131,10 +131,9 @@ pub fn try_to_bigint(value: f64, vm: &VirtualMachine) -> PyResult<BigInt> {
         Some(int) => Ok(int),
         None => {
             if value.is_infinite() {
-                Err(vm
-                    .new_overflow_error("OverflowError: cannot convert float infinity to integer"))
+                Err(vm.new_overflow_error("cannot convert float infinity to integer"))
             } else if value.is_nan() {
-                Err(vm.new_value_error("ValueError: cannot convert float NaN to integer"))
+                Err(vm.new_value_error("cannot convert float NaN to integer"))
             } else {
                 // unreachable unless BigInt has a bug
                 unreachable!(
diff --git a/extra_tests/snippets/builtin_float.py b/extra_tests/snippets/builtin_float.py
index fa5a734fd..f0fcae5d1 100644
--- a/extra_tests/snippets/builtin_float.py
+++ b/extra_tests/snippets/builtin_float.py
@@ -517,3 +517,35 @@ assert_raises(ValueError, lambda: float("._0"))
 
 assert 3.0 % -2.0 == -1.0
 assert -3.0 % 2.0 == 1.0
+
+
+# Float-to-int conversion error wording matches CPython exactly. The
+# error class name is added by Python's exception display layer; the
+# message itself must not duplicate it.
+def _check_msg(call, exc_type, expected_msg):
+    try:
+        call()
+    except exc_type as e:
+        assert str(e) == expected_msg, repr(e)
+    else:
+        raise AssertionError(f"expected {exc_type.__name__}")
+
+
+_check_msg(lambda: int(INF), OverflowError, "cannot convert float infinity to integer")
+_check_msg(lambda: int(NINF), OverflowError, "cannot convert float infinity to integer")
+_check_msg(lambda: int(NAN), ValueError, "cannot convert float NaN to integer")
+_check_msg(
+    lambda: round(INF), OverflowError, "cannot convert float infinity to integer"
+)
+_check_msg(lambda: round(NAN), ValueError, "cannot convert float NaN to integer")
+_check_msg(
+    lambda: math.floor(INF), OverflowError, "cannot convert float infinity to integer"
+)
+_check_msg(lambda: math.ceil(NAN), ValueError, "cannot convert float NaN to integer")
+_check_msg(
+    lambda: math.trunc(INF), OverflowError, "cannot convert float infinity to integer"
+)
+_check_msg(
+    lambda: INF.__int__(), OverflowError, "cannot convert float infinity to integer"
+)
+_check_msg(lambda: NAN.__floor__(), ValueError, "cannot convert float NaN to integer")