feat(proxy): add model_group filter to /spend/logs/v2 endpoint

Add an optional `model_group` query parameter to the `/spend/logs/v2`
and `/spend/logs/ui` endpoints, allowing users to filter spend logs by
model group. This is consistent with the existing `model` and `model_id`
filters and requires no schema changes since `model_group` is already a
column in the `LiteLLM_SpendLogs` table.

Fixes #24781

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

litellm/proxy/spend_tracking/spend_management_endpoints.py

@@ -1740,6 +1740,9 @@ async def ui_view_spend_logs(  # noqa: PLR0915
         default=None,
         description="Filter logs by model ID (litellm model deployment id)",
     ),
+    model_group: Optional[str] = fastapi.Query(
+        default=None, description="Filter logs by model group"
+    ),
     key_alias: Optional[str] = fastapi.Query(
         default=None, description="Filter logs by key alias"
     ),
@@ -1869,6 +1872,9 @@ async def ui_view_spend_logs(  # noqa: PLR0915
         if model_id is not None:
             where_conditions["model_id"] = model_id
 
+        if model_group is not None:
+            where_conditions["model_group"] = model_group
+
         # Build metadata filters
         metadata_filters = []
         if key_alias is not None:
@@ -1992,6 +1998,7 @@ async def ui_view_spend_logs(  # noqa: PLR0915
             ("request_id", "request_id"),
             ("model", "model"),
             ("model_id", "model_id"),
+            ("model_group", "model_group"),
             ("end_user", "end_user"),
         ]:
             val = where_conditions.get(wc_key)

tests/test_litellm/proxy/spend_tracking/test_spend_management_endpoints.py

@@ -1373,6 +1373,71 @@ async def test_ui_view_spend_logs_with_model_id(client, monkeypatch):
         app.dependency_overrides.pop(ps.user_api_key_auth, None)
 
 
+@pytest.mark.asyncio
+async def test_ui_view_spend_logs_with_model_group(client, monkeypatch):
+    """Test that the model_group query param filters spend logs by model group."""
+    mock_spend_logs = [
+        {
+            "id": "log1",
+            "request_id": "req1",
+            "api_key": "sk-test-key",
+            "user": "test_user_1",
+            "team_id": "team1",
+            "spend": 0.05,
+            "startTime": datetime.datetime.now(timezone.utc).isoformat(),
+            "model": "gpt-3.5-turbo",
+            "model_group": "gpt-3.5-turbo",
+            "status": "success",
+        },
+        {
+            "id": "log2",
+            "request_id": "req2",
+            "api_key": "sk-test-key",
+            "user": "test_user_2",
+            "team_id": "team1",
+            "spend": 0.10,
+            "startTime": datetime.datetime.now(timezone.utc).isoformat(),
+            "model": "gpt-4-0613",
+            "model_group": "gpt-4",
+            "status": "success",
+        },
+    ]
+
+    def filter_by_model_group(where):
+        if "model_group" in where and where["model_group"] == "gpt-4":
+            return [mock_spend_logs[1]]
+        return mock_spend_logs
+
+    monkeypatch.setattr(
+        "litellm.proxy.proxy_server.prisma_client",
+        make_ui_spend_logs_mock_prisma(mock_spend_logs, filter_by_model_group),
+    )
+
+    start_date, end_date = _default_date_range()
+
+    app.dependency_overrides[ps.user_api_key_auth] = lambda: UserAPIKeyAuth(
+        user_role=LitellmUserRoles.PROXY_ADMIN
+    )
+    try:
+        response = client.get(
+            "/spend/logs/ui",
+            params={
+                "model_group": "gpt-4",
+                "start_date": start_date,
+                "end_date": end_date,
+            },
+            headers={"Authorization": "Bearer sk-test"},
+        )
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["total"] == 1
+        assert len(data["data"]) == 1
+        assert data["data"][0]["model_group"] == "gpt-4"
+    finally:
+        app.dependency_overrides.pop(ps.user_api_key_auth, None)
+
+
 @pytest.mark.asyncio
 async def test_ui_view_spend_logs_with_key_hash(client, monkeypatch):
     mock_spend_logs = [
@@ -1465,10 +1530,13 @@ class TestSpendLogsPayload:
         litellm.callbacks = [_ProxyDBLogger(message_logging=False)]
         # litellm._turn_on_debug()
 
-        with patch.object(
-            litellm.proxy.db.db_spend_update_writer.DBSpendUpdateWriter,
-            "_insert_spend_log_to_db",
-        ) as mock_client, patch.object(litellm.proxy.proxy_server, "prisma_client"):
+        with (
+            patch.object(
+                litellm.proxy.db.db_spend_update_writer.DBSpendUpdateWriter,
+                "_insert_spend_log_to_db",
+            ) as mock_client,
+            patch.object(litellm.proxy.proxy_server, "prisma_client"),
+        ):
             response = await litellm.acompletion(
                 model="gpt-4o",
                 messages=[{"role": "user", "content": "Hello, world!"}],
@@ -1558,13 +1626,13 @@ class TestSpendLogsPayload:
 
         client = AsyncHTTPHandler()
 
-        with patch.object(
-            litellm.proxy.db.db_spend_update_writer.DBSpendUpdateWriter,
-            "_insert_spend_log_to_db",
-        ) as mock_client, patch.object(
-            litellm.proxy.proxy_server, "prisma_client"
-        ), patch.object(
-            client, "post", side_effect=self.mock_anthropic_response
+        with (
+            patch.object(
+                litellm.proxy.db.db_spend_update_writer.DBSpendUpdateWriter,
+                "_insert_spend_log_to_db",
+            ) as mock_client,
+            patch.object(litellm.proxy.proxy_server, "prisma_client"),
+            patch.object(client, "post", side_effect=self.mock_anthropic_response),
         ):
             response = await litellm.acompletion(
                 model="claude-4-sonnet-20250514",
@@ -1652,13 +1720,13 @@ class TestSpendLogsPayload:
             ]
         )
 
-        with patch.object(
-            litellm.proxy.db.db_spend_update_writer.DBSpendUpdateWriter,
-            "_insert_spend_log_to_db",
-        ) as mock_client, patch.object(
-            litellm.proxy.proxy_server, "prisma_client"
-        ), patch.object(
-            client, "post", side_effect=self.mock_anthropic_response
+        with (
+            patch.object(
+                litellm.proxy.db.db_spend_update_writer.DBSpendUpdateWriter,
+                "_insert_spend_log_to_db",
+            ) as mock_client,
+            patch.object(litellm.proxy.proxy_server, "prisma_client"),
+            patch.object(client, "post", side_effect=self.mock_anthropic_response),
         ):
             response = await router.acompletion(
                 model="my-anthropic-model-group",

uv.lock

@@ -11,7 +11,7 @@ resolution-markers = [
 ]
 
 [options]
-exclude-newer = "2026-04-08T16:01:27.663665Z"
+exclude-newer = "2026-04-12T07:48:54.018235Z"
 exclude-newer-span = "P3D"
 
 [manifest]
@@ -3602,7 +3602,7 @@ wheels = [
 
 [[package]]
 name = "litellm"
-version = "1.83.6"
+version = "1.83.8"
 source = { editable = "." }
 dependencies = [
     { name = "aiohttp" },