diff --git a/django/db/models/query.py b/django/db/models/query.py
index a4277d05fc..a0a0572728 100644
--- a/django/db/models/query.py
+++ b/django/db/models/query.py
@@ -1119,6 +1119,29 @@ class QuerySet(AltersData):
                 "in_bulk()'s field_name must be a unique field but %r isn't."
                 % field_name
             )
+
+        if issubclass(self._iterable_class, ModelIterable):
+
+            def _get_key(row):
+                # Will raise an unhelpful AttributeError if field_name is deferred
+                # We could catch this earlier and raise a TypeError
+                return getattr(row, field_name)
+
+        elif issubclass(self._iterable_class, ValuesIterable):
+            # If field_name wasn't explicitly included, include it anyways
+            if field_name not in self.query.values_select:
+                return self.values(field_name, *self.query.values_select).in_bulk(
+                    id_list=id_list, field_name=field_name
+                )
+
+            def _get_key(row):
+                return row[field_name]
+
+        else:
+            # NamedValuesListIterable should work in theory,
+            # but in practice it's very clunky.
+            raise TypeError("in_bulk() cannot be used with %r." % self._iterable_class)
+
         if id_list is not None:
             if not id_list:
                 return {}
@@ -1136,7 +1159,7 @@ class QuerySet(AltersData):
                 qs = self.filter(**{filter_key: id_list})
         else:
             qs = self._chain()
-        return {getattr(obj, field_name): obj for obj in qs}
+        return {_get_key(obj): obj for obj in qs}
 
     async def ain_bulk(self, id_list=None, *, field_name="pk"):
         return await sync_to_async(self.in_bulk)(
diff --git a/tests/lookup/tests.py b/tests/lookup/tests.py
index df96546d04..8f04537ec1 100644
--- a/tests/lookup/tests.py
+++ b/tests/lookup/tests.py
@@ -205,6 +205,23 @@ class LookupTests(TestCase):
         with self.assertRaises(TypeError):
             Article.objects.in_bulk(headline__startswith="Blah")
 
+    def test_in_bulk_values(self):
+        bulk = Article.objects.values().in_bulk([self.a1.id])
+        self.assertIsInstance(bulk[self.a1.id], dict)
+
+    def test_in_bulk_values_without_pk(self):
+        bulk = Article.objects.values("headline").in_bulk([self.a1.id])
+        self.assertIn("pk", bulk[self.a1.id])
+
+    def test_in_bulk_values_list(self):
+        with self.assertRaises(TypeError):
+            Article.objects.values_list().in_bulk([self.a1.id])
+
+    def test_in_bulk_values_list_named(self) -> None:
+        # At some point this could return tuples
+        with self.assertRaises(TypeError):
+            Article.objects.values_list(named=True).in_bulk([self.a1.id])
+
     def test_in_bulk_lots_of_ids(self):
         test_range = 2000
         max_query_params = connection.features.max_query_params