diff --git a/codespeed/results.py b/codespeed/results.py
index 97e3e2ea..027f2ffd 100644
--- a/codespeed/results.py
+++ b/codespeed/results.py
@@ -32,6 +32,7 @@ def validate_result(item):
         'benchmark',
         'environment',
         'result_value',
+        'source',
     ]
 
     error = True
@@ -41,7 +42,7 @@ def validate_result(item):
         elif key in item and item[key] == "":
             return 'Value for key "' + key + '" empty in request', error
 
-    if 'source' in item and item['source'] not in dict(Benchmark.S_TYPES):
+    if item['source'] not in dict(Benchmark.S_TYPES):
         return 'Invalid source "%s"' % item['source'], error
 
     try:
@@ -73,9 +74,8 @@ def save_result(data, update_repo=True):
     p, created = Project.objects.get_or_create(name=data["project"])
     branch, created = Branch.objects.get_or_create(name=data["branch"],
                                                    project=p)
-    source = data.get("source", "legacy")
     b, created = Benchmark.objects.get_or_create(
-        name=data["benchmark"], source=source)
+        name=data["benchmark"], source=data["source"])
 
     if created:
         if "description" in data:
diff --git a/codespeed/tests/test_views.py b/codespeed/tests/test_views.py
index 8f07a61f..3108d5f4 100644
--- a/codespeed/tests/test_views.py
+++ b/codespeed/tests/test_views.py
@@ -29,6 +29,7 @@ def setUp(self):
             'benchmark': 'float',
             'environment': 'Dual Core',
             'result_value': 456,
+            'source': 'legacy',
         }
 
     def test_add_correct_result(self):
@@ -190,11 +191,13 @@ def test_source_set_on_new_benchmark(self):
         b = Benchmark.objects.get(name='newbench')
         self.assertEqual(b.source, 'pyperformance')
 
-    def test_source_defaults_to_legacy(self):
-        """A payload without a source creates a 'legacy' Benchmark"""
-        self.client.post(self.path, self.data)
-        b = Benchmark.objects.get(name='float')
-        self.assertEqual(b.source, 'legacy')
+    def test_missing_source_rejected(self):
+        """A payload without a source is rejected instead of defaulting"""
+        modified_data = copy.deepcopy(self.data)
+        del modified_data['source']
+        response = self.client.post(self.path, modified_data)
+        self.assertEqual(response.status_code, 400)
+        self.assertFalse(Benchmark.objects.filter(name='float').exists())
 
     def test_same_name_different_source_are_distinct(self):
         """The same name in a different suite is a separate Benchmark"""
@@ -245,28 +248,32 @@ def setUp(self):
              'executable': 'pypy-c',
              'benchmark': 'Richards',
              'environment': 'bigdog',
-             'result_value': 456},
+             'result_value': 456,
+             'source': 'legacy'},
             {'commitid': '456',
              'project': 'pypy',
              'branch': 'default',
              'executable': 'pypy-c',
              'benchmark': 'Richards',
              'environment': 'bigdog',
-             'result_value': 457},
+             'result_value': 457,
+             'source': 'legacy'},
             {'commitid': '456',
              'project': 'pypy',
              'branch': 'default',
              'executable': 'pypy-c',
              'benchmark': 'Richards2',
              'environment': 'bigdog',
-             'result_value': 34},
+             'result_value': 34,
+             'source': 'legacy'},
             {'commitid': '789',
              'project': 'pypy',
              'branch': 'default',
              'executable': 'pypy-c',
              'benchmark': 'Richards',
              'environment': 'bigdog',
-             'result_value': 458},
+             'result_value': 458,
+             'source': 'legacy'},
         ]
 
     def test_get_returns_405(self):
@@ -452,6 +459,7 @@ def setUp(self):
             'benchmark': 'float',
             'environment': 'Dual Core',
             'result_value': 200,
+            'source': 'legacy',
         }
         resp = self.client.post(reverse('add-result'),
                                 self.data)
diff --git a/codespeed/views.py b/codespeed/views.py
index 723f2e59..deb6d2b0 100644
--- a/codespeed/views.py
+++ b/codespeed/views.py
@@ -114,7 +114,9 @@ def embed_comparison(request):
 
 @require_GET
 def gethistoricaldata(request):
-    data = {'results': {}, 'benchmarks': []}
+    # 'labels' maps each key in tagged_revs (plus baseline and 'latest') to
+    # a two-line label [tag, project] for the x axis of the history chart.
+    data = {'results': {}, 'benchmarks': [], 'labels': {}}
     env = Environment.objects.all()
     if settings.DEF_ENVIRONMENT:
         env = env.get(name=settings.DEF_ENVIRONMENT)
@@ -134,6 +136,7 @@ def gethistoricaldata(request):
             )
         rev0 = rev[0]
         resname = '{} {}'.format(b['executable'], rev0.tag)
+        data['labels'][resname] = [rev0.tag, baseline_exe.project.name]
         baseline_results.append((resname, Result.objects.filter(
             executable=baseline_exe, revision=rev0, environment=env,
             benchmark__source='legacy')))
@@ -165,12 +168,14 @@ def gethistoricaldata(request):
                 logger.info("no results for '%s' '%s' '%s'" % (str(_default_exe), str(rev), str(env)))
                 continue
             default_results[rev.tag] = res
+            data['labels'][rev.tag] = [rev.tag, _default_exe.project.name]
     data['tagged_revs'] = [rev.tag for rev in all_taggedrevs if rev.tag in default_results]
     # Fetch data for latest results
     executable = settings.DEF_EXECUTABLES[0]
     def_name = executable['name']
     def_project = Project.objects.get(name=executable['project'])
     default_exe = Executable.objects.get(name=def_name, project=def_project)
+    data['labels']['latest'] = ['latest', default_exe.project.name]
     default_branch = Branch.objects.get(
             name=default_exe.project.default_branch,
             project=default_exe.project)
diff --git a/speed_pypy/settings.py b/speed_pypy/settings.py
index 4d6a559f..722b9c0e 100644
--- a/speed_pypy/settings.py
+++ b/speed_pypy/settings.py
@@ -93,6 +93,7 @@
                  {'executable': 'cpython', 'revision': '3.11.15'},
                 ]
 DEF_EXECUTABLES = [
+                   {'name': 'pypy3.12-jit-64', 'project': 'PyPy3.12'},
                    {'name': 'pypy3.11-jit-64', 'project': 'PyPy3.11'},
                   ]
 DEF_ENVIRONMENT = 'benchmarker2'
diff --git a/speed_pypy/templates/embed_comparison.html b/speed_pypy/templates/embed_comparison.html
index 04f30028..5bd6c565 100644
--- a/speed_pypy/templates/embed_comparison.html
+++ b/speed_pypy/templates/embed_comparison.html
@@ -6,11 +6,13 @@
   <meta name="viewport" content="width=device-width, initial-scale=1">
   <title>PyPy vs CPython benchmark comparison</title>
   <style>
-    html, body { margin: 0; padding: 0; background: transparent; }
+    html, body { margin: 0; padding: 0; background: transparent; height: 100%; overflow: hidden; }
+    /* Fill the embedding iframe so the rotated benchmark labels are never
+       clipped: the host page controls the height via the iframe. */
     #baseline-comparison-plot {
       position: relative;
       width: 100%;
-      height: 480px;
+      height: 100vh;
       font-family: sans-serif;
     }
   </style>
diff --git a/speed_pypy/templates/home.html b/speed_pypy/templates/home.html
index 01d15c97..b6cea97b 100644
--- a/speed_pypy/templates/home.html
+++ b/speed_pypy/templates/home.html
@@ -31,7 +31,7 @@ <h2>Comparison</h2>
     <h3>How fast is {{ default_exe.project }}?</h3>
     <iframe src="{% url 'embed-comparison' %}"
             title="{{ default_exe.project }} ({{ default_exe }}) vs {{ baseline }} benchmark comparison"
-            style="border:0;width:100%;height:400px"
+            style="border:0;width:100%;height:480px"
             loading="lazy"></iframe>
     <p class="plot-caption">Plot 1: The above plot represents {{ default_exe.project }} ({{ default_exe }}) benchmark times normalized to {{ baseline }}. Smaller is better.</p>
     <p>It depends greatly on the type of task being performed. The geometric average of all benchmarks is <span id="geomean"></span> or <strong id="geofaster"></strong> times <em>faster</em> than {{ baseline }}</p>
@@ -120,9 +120,11 @@ <h3>How has PyPy performance evolved over time?</h3>
         geomeans.push(1 / trunk_geomean);
         $('#num_of_benchs').html(num_of_benchs);
 
-        var ticks2 = [data['baseline']];
-        for (var i in data['tagged_revs']) { ticks2.push(data['tagged_revs'][i]); }
-        ticks2.push('latest {{ default_exe.project }}');
+        // Two-line labels: [tag, project]; fall back to the key itself.
+        var labelFor = function(key) { return (data['labels'] || {})[key] || key; };
+        var ticks2 = [labelFor(data['baseline'])];
+        for (var i in data['tagged_revs']) { ticks2.push(labelFor(data['tagged_revs'][i])); }
+        ticks2.push(labelFor('latest'));
 
         var wrap2 = document.getElementById('historical-plot');
         var canvas2 = document.createElement('canvas');
diff --git a/tools/pypy/savecpython.py b/tools/pypy/savecpython.py
index c5200947..c15ce033 100644
--- a/tools/pypy/savecpython.py
+++ b/tools/pypy/savecpython.py
@@ -31,6 +31,7 @@ def save(project, revision, results, options, executable, host, testing=False):
             'environment': host,
             'result_value': value,
             'result_date': current_date,
+            'source': 'legacy',
         }
         if res_type == "ComparisonResult":
             data['std_dev'] = results['std_changed']
diff --git a/tools/pypy/saveresults.py b/tools/pypy/saveresults.py
index 3d3eb0df..ddc000e5 100644
--- a/tools/pypy/saveresults.py
+++ b/tools/pypy/saveresults.py
@@ -79,6 +79,7 @@ def save(project, revision, results, executable, host, url, testing=False,
             'environment': host,
             'result_value': value,
             'branch': branch,
+            'source': 'legacy',
         }]
         if not value:
             print("Ignoring skipped result", data)
diff --git a/tools/save_multiple_results.py b/tools/save_multiple_results.py
index bc262e1a..82d340b7 100644
--- a/tools/save_multiple_results.py
+++ b/tools/save_multiple_results.py
@@ -18,7 +18,8 @@
         "executable": "myexe O3 64bits",
         "benchmark": "float",
         "environment": "Dual Core",
-        "result_value": 2500.0
+        "result_value": 2500.0,
+        "source": "legacy"
     },
     {
         "commitid": "8",
@@ -27,7 +28,8 @@
         "executable": "myexe O3 64bits",
         "benchmark": "int",
         "environment": "Dual Core",
-        "result_value": 1100
+        "result_value": 1100,
+        "source": "legacy"
     }
 ]
 
diff --git a/tools/save_single_result.py b/tools/save_single_result.py
index 830858e7..736cb77e 100644
--- a/tools/save_single_result.py
+++ b/tools/save_single_result.py
@@ -20,6 +20,7 @@
     'benchmark': 'float',
     'environment': "Dual Core",
     'result_value': 4000,
+    'source': 'legacy',  # 'legacy' or 'pyperformance'
 }
 
 # Optional fields