diff --git a/Doc/library/doctest.rst b/Doc/library/doctest.rst
index 3298697af8511b6..4b61c07919dd025 100644
--- a/Doc/library/doctest.rst
+++ b/Doc/library/doctest.rst
@@ -1162,6 +1162,12 @@ from text files and modules with doctests:
    .. versionchanged:: 3.15
       Run each example as a :ref:`subtest <subtests>`.
 
+   .. versionchanged:: next
+      Report every example, as in verbose mode, if the test runner reports
+      more than the test names, i.e. its
+      :attr:`~unittest.TestResult.verbosity` is 3 or higher (for example
+      with ``python -m unittest -vv``).
+
 Under the covers, :func:`DocTestSuite` creates a :class:`unittest.TestSuite` out
 of :class:`!doctest.DocTestCase` instances, and :class:`!DocTestCase` is a
 subclass of :class:`unittest.TestCase`. :class:`!DocTestCase` isn't documented
diff --git a/Doc/library/test.rst b/Doc/library/test.rst
index f3b5383658b5ac4..ba6600fc46679f9 100644
--- a/Doc/library/test.rst
+++ b/Doc/library/test.rst
@@ -232,9 +232,10 @@ The :mod:`!test.support` module defines the following constants:
 
 .. data:: verbose
 
-   ``True`` when verbose output is enabled. Should be checked when more
-   detailed information is desired about a running test. *verbose* is set by
-   :mod:`test.regrtest`.
+   How verbose the output is: the number of :option:`!-v` options which
+   :mod:`test.regrtest` was run with, and therefore ``0`` when verbose output
+   is not enabled.  Should be checked when more detailed information is
+   desired about a running test.
 
 
 .. data:: is_jython
diff --git a/Doc/library/unittest.rst b/Doc/library/unittest.rst
index 7afcdb368a3562e..e1bc32fb79c2eb1 100644
--- a/Doc/library/unittest.rst
+++ b/Doc/library/unittest.rst
@@ -180,6 +180,9 @@ You can run tests with more detail (higher verbosity) by passing in the -v flag:
 
    python -m unittest -v test_module
 
+Repeat it for even more detail: ``-vv`` reports also the individual examples
+of a :mod:`doctest`.
+
 When executed without arguments :ref:`unittest-test-discovery` is started::
 
    python -m unittest
@@ -291,7 +294,11 @@ The ``discover`` sub-command has the following options:
 
 .. option:: -v, --verbose
 
-   Verbose output
+   Verbose output.  May be repeated: ``-vv`` reports also the individual
+   examples of a :mod:`doctest`.
+
+   .. versionchanged:: next
+      The option can be repeated.
 
 .. option:: -s, --start-directory directory
 
@@ -2149,6 +2156,15 @@ Loading and running tests
 
       .. versionadded:: 3.5
 
+   .. attribute:: verbosity
+
+      The level of details which the test runner reports: ``0`` -- quiet,
+      ``1`` -- progress dots, ``2`` -- test names, ``3`` -- also the
+      individual examples of a :mod:`doctest`.  A test runner is expected to
+      set it to its own verbosity.
+
+      .. versionadded:: next
+
    .. method:: wasSuccessful()
 
       Return ``True`` if all tests run so far have passed, otherwise returns
diff --git a/Lib/doctest.py b/Lib/doctest.py
index 8a55fe3ddd26154..dceea82f54bcd79 100644
--- a/Lib/doctest.py
+++ b/Lib/doctest.py
@@ -1200,6 +1200,18 @@ def _find_lineno(self, obj, source_lines):
 ## 5. DocTest Runner
 ######################################################################
 
+def _make_output_function(stream):
+    """Return a function writing to *stream*, whatever it can encode."""
+    encoding = getattr(stream, 'encoding', None)
+    if encoding is None or encoding.lower() == 'utf-8':
+        return stream.write
+    def out(s):
+        # Use backslashreplace error handling on write
+        s = str(s.encode(encoding, 'backslashreplace'), encoding)
+        stream.write(s)
+    return out
+
+
 class DocTestRunner:
     """
     A class used to run DocTest test cases, and accumulate statistics.
@@ -1560,14 +1572,7 @@ def run(self, test, compileflags=None, out=None, clear_globs=True):
 
         save_stdout = sys.stdout
         if out is None:
-            encoding = save_stdout.encoding
-            if encoding is None or encoding.lower() == 'utf-8':
-                out = save_stdout.write
-            else:
-                # Use backslashreplace error handling on write
-                def out(s):
-                    s = str(s.encode(encoding, 'backslashreplace'), encoding)
-                    save_stdout.write(s)
+            out = _make_output_function(save_stdout)
         sys.stdout = self._fakeout
 
         # Patch pdb.set_trace to restore sys.stdout during interactive
@@ -2321,6 +2326,9 @@ def report_skip(self, out, test, example):
         unittest.case._addSkip(self._test_result, self._subTest(), '')
 
     def report_success(self, out, test, example, got):
+        # Report "ok" if verbose, to close what report_start() opened.  A
+        # failed or skipped example is reported by the test result instead.
+        super().report_success(out, test, example, got)
         self._test_result.addSubTest(self._test_case, self._subTest(), None)
 
     def report_unexpected_exception(self, out, test, example, exc_info):
@@ -2400,10 +2408,23 @@ def runTest(self):
         if getattr(result, 'failfast', False):
             optionflags |= FAIL_FAST
 
+        # Report every example only if the test runner is asked for more than
+        # the test names it reports at verbosity 2.  Write them to its stream,
+        # so that they are not swallowed by result.buffer.
+        verbose = getattr(result, 'verbosity', 1) >= 3
+        stream = getattr(result, 'stream', None)
+        out = None
+        if verbose and stream is not None:
+            out = _make_output_function(stream)
+            if test.examples and not getattr(result, '_newline', True):
+                # End the line which startTest() left open.
+                out('\n')
+                result._newline = True
+
         runner = _DocTestCaseRunner(optionflags=optionflags,
-                               checker=self._dt_checker, verbose=False,
+                               checker=self._dt_checker, verbose=verbose,
                                test_case=self, test_result=result)
-        results = runner.run(test, clear_globs=False)
+        results = runner.run(test, out=out, clear_globs=False)
         if results.skipped == results.attempted:
             raise unittest.SkipTest("all examples were skipped")
 
diff --git a/Lib/test/libregrtest/testresult.py b/Lib/test/libregrtest/testresult.py
index 1820f354572521a..605f1f4e6a89fb6 100644
--- a/Lib/test/libregrtest/testresult.py
+++ b/Lib/test/libregrtest/testresult.py
@@ -16,7 +16,7 @@ class RegressionTestResult(unittest.TextTestResult):
 
     def __init__(self, stream, descriptions, verbosity):
         super().__init__(stream=stream, descriptions=descriptions,
-                         verbosity=2 if verbosity else 0)
+                         verbosity=verbosity)
         self.buffer = True
         if self.USE_XML:
             from xml.etree import ElementTree as ET
@@ -150,10 +150,12 @@ def run(self, test):
 
 def get_test_runner_class(verbosity, buffer=False):
     if verbosity:
+        # The verbosity of regrtest is one less than the verbosity of
+        # unittest: -v reports the test names, -vv also the doctest examples.
         return functools.partial(unittest.TextTestRunner,
                                  resultclass=RegressionTestResult,
                                  buffer=buffer,
-                                 verbosity=verbosity)
+                                 verbosity=verbosity + 1)
     return functools.partial(QuietRegressionTestRunner, buffer=buffer)
 
 def get_test_runner(stream, verbosity, capture_output=False):
diff --git a/Lib/test/test_doctest/test_doctest.py b/Lib/test/test_doctest/test_doctest.py
index b125693ab0891cd..776ad83ee6d6ce5 100644
--- a/Lib/test/test_doctest/test_doctest.py
+++ b/Lib/test/test_doctest/test_doctest.py
@@ -6,6 +6,7 @@
 from test.support import import_helper
 import doctest
 import functools
+import io
 import os
 import sys
 import importlib
@@ -469,7 +470,7 @@ def basics(): r"""
     >>> tests = finder.find(sample_func)
 
     >>> print(tests)  # doctest: +ELLIPSIS
-    [<DocTest sample_func from test_doctest.py:36 (1 example)>]
+    [<DocTest sample_func from test_doctest.py:37 (1 example)>]
 
 The exact name depends on how test_doctest was invoked, so allow for
 leading path components.
@@ -803,6 +804,59 @@ def myfunc():
         self.assertEqual((x, y), (2, 3))
 
 
+class TestDocTestSuiteVerbosity(unittest.TestCase):
+
+    def run_suite(self, module='test.test_doctest.sample_doctest', **kwargs):
+        """Return what the test runner wrote and what leaked to stdout."""
+        suite = doctest.DocTestSuite(module)
+        stream = io.StringIO()
+        stdout = io.StringIO()
+        with contextlib.redirect_stdout(stdout):
+            unittest.TextTestRunner(stream=stream, **kwargs).run(suite)
+        return stream.getvalue(), stdout.getvalue()
+
+    def test_quiet(self):
+        for verbosity in range(3):
+            with self.subTest(verbosity=verbosity):
+                output, stdout = self.run_suite(verbosity=verbosity)
+                self.assertNotIn('Trying:', output)
+                self.assertNotIn('Expecting:', output)
+                self.assertEqual(stdout, '')
+
+    def test_verbose(self):
+        output, stdout = self.run_suite(verbosity=3)
+        self.assertIn('Trying:\n    2+2\n', output)
+        self.assertIn('Expecting:\n    4\n', output)
+        self.assertIn('\nok\n', output)
+        # Reported to the stream of the test runner, not to the stdout.
+        self.assertEqual(stdout, '')
+
+    def test_verbose_buffered(self):
+        # result.buffer replaces sys.stdout, which would swallow the examples.
+        output, stdout = self.run_suite(verbosity=3, buffer=True)
+        self.assertIn('Trying:\n    2+2\n', output)
+        self.assertEqual(stdout, '')
+
+    def test_verbose_failure_not_duplicated(self):
+        module = 'test.test_doctest.sample_doctest_errors'
+        quiet, _ = self.run_suite(module, verbosity=2)
+        verbose, _ = self.run_suite(module, verbosity=3)
+        self.assertIn('Trying:', verbose)
+        self.assertNotIn('Trying:', quiet)
+        # Reporting the examples does not report the failures once more.
+        self.assertEqual(verbose.count('Failed example:'),
+                         quiet.count('Failed example:'))
+        self.assertGreater(quiet.count('Failed example:'), 0)
+
+    def test_plain_result(self):
+        # A result which is not from a text test runner has no stream.
+        suite = doctest.DocTestSuite('test.test_doctest.sample_doctest')
+        stdout = io.StringIO()
+        with contextlib.redirect_stdout(stdout):
+            suite.run(unittest.TestResult())
+        self.assertEqual(stdout.getvalue(), '')
+
+
 class TestDocTestFinder(unittest.TestCase):
 
     def test_issue35753(self):
diff --git a/Lib/test/test_regrtest.py b/Lib/test/test_regrtest.py
index 6ba440053089161..1245c229722f04b 100644
--- a/Lib/test/test_regrtest.py
+++ b/Lib/test/test_regrtest.py
@@ -2151,6 +2151,18 @@ def load_tests(loader, tests, pattern):
                                   failed=[testname],
                                   parallel=True,
                                   stats=TestStats(1, 2, 1))
+        # A single -v reports the test names, not the examples.
+        self.assertNotIn('Trying:', output)
+
+        # -vv reports every example, without changing what is run.
+        output = self.run_tests("--fail-env-changed", "-vv", "-j1", testname,
+                                exitcode=EXITCODE_BAD_TEST)
+        self.check_executed_tests(output, [testname],
+                                  failed=[testname],
+                                  parallel=True,
+                                  stats=TestStats(1, 2, 1))
+        self.assertIn('Trying:\n    1 + 1\n', output)
+        self.assertIn('Expecting:\n    2\n', output)
 
     def _check_random_seed(self, run_workers: bool):
         # gh-109276: When -r/--randomize is used, random.seed() is called
diff --git a/Lib/test/test_unittest/test_discovery.py b/Lib/test/test_unittest/test_discovery.py
index 38c9779daaf87d4..da184bd12be8d17 100644
--- a/Lib/test/test_unittest/test_discovery.py
+++ b/Lib/test/test_unittest/test_discovery.py
@@ -635,7 +635,7 @@ def test_command_line_handling_discover_by_default_with_options(self):
         program._do_discovery = args.append
         program.parseArgs(['something', '-v', '-b', '-v', '-c', '-f'])
         self.assertEqual(args, [[]])
-        self.assertEqual(program.verbosity, 2)
+        self.assertEqual(program.verbosity, 3)  # -v is passed twice
         self.assertIs(program.buffer, True)
         self.assertIs(program.catchbreak, True)
         self.assertIs(program.failfast, True)
diff --git a/Lib/test/test_unittest/test_program.py b/Lib/test/test_unittest/test_program.py
index 8ed92373e5e984a..a9a73f0d057ee81 100644
--- a/Lib/test/test_unittest/test_program.py
+++ b/Lib/test/test_unittest/test_program.py
@@ -267,6 +267,32 @@ def testVerbosity(self):
             program.parseArgs([None, opt])
             self.assertEqual(program.verbosity, 2)
 
+        # -v can be repeated to ask for more details.
+        for args, verbosity in (
+            (['-vv'], 3),
+            (['-v', '-v'], 3),
+            (['--verbose', '--verbose'], 3),
+            (['-vvv'], 4),
+            # -q overrides any number of -v.
+            (['-v', '-q'], 0),
+        ):
+            with self.subTest(args=args):
+                program.verbosity = 1
+                program.parseArgs([None, *args])
+                self.assertEqual(program.verbosity, verbosity)
+
+    def testVerbosityCountedOnce(self):
+        # "python -m unittest -v" falls back to test discovery, which parses
+        # arguments again: -v must not be counted twice.
+        program = self.program
+        program.verbosity = 1
+        program.parseArgs([None, '-v'])
+        self.assertEqual(program.verbosity, 2)
+
+        program.verbosity = 1
+        program.parseArgs([None, 'discover', '-vv'])
+        self.assertEqual(program.verbosity, 3)
+
     def testBufferCatchFailfast(self):
         program = self.program
         for arg, attr in (('buffer', 'buffer'), ('failfast', 'failfast'),
diff --git a/Lib/test/test_unittest/test_result.py b/Lib/test/test_unittest/test_result.py
index 3f44e617303f81d..cb591c23fc964fd 100644
--- a/Lib/test/test_unittest/test_result.py
+++ b/Lib/test/test_unittest/test_result.py
@@ -54,6 +54,13 @@ def test_init(self):
         self.assertEqual(result.shouldStop, False)
         self.assertIsNone(result._stdout_buffer)
         self.assertIsNone(result._stderr_buffer)
+        self.assertEqual(result.verbosity, 1)
+
+    def test_init_verbosity(self):
+        for verbosity in range(4):
+            with self.subTest(verbosity=verbosity):
+                result = unittest.TestResult(None, None, verbosity)
+                self.assertEqual(result.verbosity, verbosity)
 
     # "This method can be called to signal that the set of tests being
     # run should be aborted by setting the TestResult's shouldStop
diff --git a/Lib/test/test_unittest/test_runner.py b/Lib/test/test_unittest/test_runner.py
index a47e2ebb59da028..195d5e928030927 100644
--- a/Lib/test/test_unittest/test_runner.py
+++ b/Lib/test/test_unittest/test_runner.py
@@ -1363,6 +1363,35 @@ def MockResultClass(*args):
         expectedresult = (runner.stream, DESCRIPTIONS, VERBOSITY)
         self.assertEqual(runner._makeResult(), expectedresult)
 
+    def test_verbosity_set_on_result(self):
+        class Suite:
+            def __call__(self, result):
+                pass
+
+        for verbosity in range(4):
+            with self.subTest(verbosity=verbosity):
+                runner = unittest.TextTestRunner(io.StringIO(),
+                                                 verbosity=verbosity)
+                result = runner.run(Suite())
+                self.assertEqual(result.verbosity, verbosity)
+
+    def test_verbosity_set_on_filtering_result(self):
+        # A result class is free to filter the verbosity which its
+        # constructor gets, as test.libregrtest does.
+        class FilteringResult(unittest.TextTestResult):
+            def __init__(self, stream, descriptions, verbosity):
+                super().__init__(stream, descriptions,
+                                 2 if verbosity else 0)
+
+        class Suite:
+            def __call__(self, result):
+                pass
+
+        runner = unittest.TextTestRunner(io.StringIO(), verbosity=3,
+                                         resultclass=FilteringResult)
+        result = runner.run(Suite())
+        self.assertEqual(result.verbosity, 3)
+
     @support.force_not_colorized
     @support.requires_subprocess()
     def test_warnings(self):
diff --git a/Lib/unittest/main.py b/Lib/unittest/main.py
index 6eeebf9657a3c7b..850c825a7366579 100644
--- a/Lib/unittest/main.py
+++ b/Lib/unittest/main.py
@@ -161,9 +161,12 @@ def _initArgParsers(self):
     def _getParentArgParser(self):
         parser = argparse.ArgumentParser(add_help=False)
 
+        # Counted, not a constant: the namespace is the TestProgram, whose
+        # verbosity is already 1, so -v still gives 2 and -vv gives 3.
         parser.add_argument('-v', '--verbose', dest='verbosity',
-                            action='store_const', const=2,
-                            help='Verbose output')
+                            action='count', default=1,
+                            help='Verbose output, twice to also report '
+                                 'the examples of a doctest')
         parser.add_argument('-q', '--quiet', dest='verbosity',
                             action='store_const', const=0,
                             help='Quiet output')
diff --git a/Lib/unittest/result.py b/Lib/unittest/result.py
index b8ea396db6772ed..a787686a5bea4ce 100644
--- a/Lib/unittest/result.py
+++ b/Lib/unittest/result.py
@@ -37,6 +37,9 @@ class TestResult(object):
     _moduleSetUpFailed = False
     def __init__(self, stream=None, descriptions=None, verbosity=None):
         self.failfast = False
+        # How much the test runner reports: 0 -- quiet, 1 -- progress dots,
+        # 2 -- test names, 3 -- also the examples of a doctest.
+        self.verbosity = 1 if verbosity is None else verbosity
         self.failures = []
         self.errors = []
         self.testsRun = 0
diff --git a/Lib/unittest/runner.py b/Lib/unittest/runner.py
index 893fcba968c3ef8..f19d7b6e446d2ac 100644
--- a/Lib/unittest/runner.py
+++ b/Lib/unittest/runner.py
@@ -244,6 +244,9 @@ def run(self, test):
         result.failfast = self.failfast
         result.buffer = self.buffer
         result.tb_locals = self.tb_locals
+        # Not left to _makeResult(): a result class is free to filter the
+        # verbosity which its constructor gets.
+        result.verbosity = self.verbosity
         with warnings.catch_warnings():
             if self.warnings:
                 # if self.warnings is set, use it to filter all the warnings
diff --git a/Misc/NEWS.d/next/Library/2026-08-08-09-14-22.gh-issue-108885.C4ktfw.rst b/Misc/NEWS.d/next/Library/2026-08-08-09-14-22.gh-issue-108885.C4ktfw.rst
new file mode 100644
index 000000000000000..e504f9b7d684f88
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2026-08-08-09-14-22.gh-issue-108885.C4ktfw.rst
@@ -0,0 +1,6 @@
+Doctests run by the :mod:`unittest` test runner now report every example, as
+in verbose mode, if the runner reports more than the test names, i.e. its
+verbosity is 3 or higher.  The :option:`!-v` option of :mod:`unittest` can now
+be repeated, so ``python -m unittest -vv`` asks for this.  Added also the
+:attr:`~unittest.TestResult.verbosity` attribute of
+:class:`unittest.TestResult`.
diff --git a/Misc/NEWS.d/next/Tests/2026-08-08-09-15-08.gh-issue-108885.Qm3vLp.rst b/Misc/NEWS.d/next/Tests/2026-08-08-09-15-08.gh-issue-108885.Qm3vLp.rst
new file mode 100644
index 000000000000000..3cdac196dd35c0c
--- /dev/null
+++ b/Misc/NEWS.d/next/Tests/2026-08-08-09-15-08.gh-issue-108885.Qm3vLp.rst
@@ -0,0 +1,2 @@
+Running the Python test suite with ``-vv`` now reports every example of a
+doctest.  A single ``-v`` reports the test names, as before.