fix(e2e): point UI tests at dashboard service; register complexity router

Stage gateway 404s /ui; the Next.js dashboard is litellm-ui:3000. Drive
playwright against E2E_UI_BASE_URL and wait on login placeholders after
client render. Register complexity-smart-router via /model/new when the
proxy does not already list it so stage matches compose config

tests/e2e/e2e_config.py

@@ -24,6 +24,11 @@ CONTROL_PLANE_BASE_URL = os.environ.get(
 UI_USERNAME = os.environ.get("E2E_UI_USERNAME", "admin")
 UI_PASSWORD = os.environ.get("E2E_UI_PASSWORD", MASTER_KEY)
 
+# Dashboard base URL. On split deployments the Next.js Admin UI is a separate
+# service (e.g. litellm-ui:3000); the data-plane gateway often 404s /ui.
+# Defaults to PROXY_BASE_URL so compose/monolith (proxy-served /ui) still works.
+UI_BASE_URL = os.environ.get("E2E_UI_BASE_URL", PROXY_BASE_URL).rstrip("/")
+
 CHEAP_ANTHROPIC_MODEL = os.environ.get("E2E_CHEAP_ANTHROPIC_MODEL", "claude-haiku-4-5")
 CHEAP_OPENAI_MODEL = os.environ.get("E2E_CHEAP_OPENAI_MODEL", "gpt-5.5")
 

tests/e2e/management/conftest.py

@@ -12,7 +12,7 @@ from typing import TYPE_CHECKING, Iterator
 
 import pytest
 
-from e2e_config import PROXY_BASE_URL, UI_PASSWORD, UI_USERNAME
+from e2e_config import UI_BASE_URL, UI_PASSWORD, UI_USERNAME
 from management_client import ManagementClient, build_client
 
 if TYPE_CHECKING:
@@ -47,13 +47,17 @@ def ui_page(browser: "Browser") -> "Iterator[Page]":
     context = browser.new_context()
     try:
         page = context.new_page()
-        page.goto(f"{PROXY_BASE_URL}/ui/")
-        page.fill("#username", UI_USERNAME)
-        page.fill("#password", UI_PASSWORD)
-        page.click('button[type="submit"]')
-        page.wait_for_function(
-            "() => document.cookie.includes('token=') || !document.querySelector('#username')"
-        )
+        # Split deploys serve the Next.js dashboard on the UI service, not the
+        # data-plane gateway (which 404s /ui). Login is a client-rendered form
+        # that appears after LoadingScreen; wait on the placeholder, not #id
+        # (Ant Design Input does not always set id="username").
+        page.goto(f"{UI_BASE_URL}/ui/login")
+        username = page.get_by_placeholder("Enter your username")
+        username.wait_for(state="visible", timeout=30_000)
+        username.fill(UI_USERNAME)
+        page.get_by_placeholder("Enter your password").fill(UI_PASSWORD)
+        page.get_by_role("button", name="Login").click()
+        page.wait_for_function("() => document.cookie.includes('token=')")
         yield page
     finally:
         context.close()

tests/e2e/management/test_key_models_dropdown_e2e.py

@@ -14,7 +14,7 @@ proxy under test does not serve it.
 
 import pytest
 
-from e2e_config import PROXY_BASE_URL, unique_marker
+from e2e_config import UI_BASE_URL, unique_marker
 from lifecycle import ResourceManager
 from management_client import ManagementClient
 from models import KeyGenerateBody, TeamNewBody
@@ -46,7 +46,7 @@ def _models_dropdown_texts(page: Page, must_contain: str) -> list[str]:
 
 
 def _open_create_key_modal(page: Page) -> None:
-    page.goto(f"{PROXY_BASE_URL}/ui/api-keys/?create=true")
+    page.goto(f"{UI_BASE_URL}/ui/api-keys/?create=true")
     expect(page.locator(".ant-modal").first).to_be_visible()
 
 
@@ -69,7 +69,7 @@ def _submit_create_modal(page: Page, sentinel_label: str) -> str:
 
 
 def _open_key_edit_form(page: Page, key_alias: str) -> None:
-    page.goto(f"{PROXY_BASE_URL}/ui/api-keys/")
+    page.goto(f"{UI_BASE_URL}/ui/api-keys/")
     page.get_by_text(key_alias).first.click()
     page.get_by_role("tab", name="Settings").click()
     page.get_by_role("button", name="Edit Settings").click()

tests/e2e/models.py

@@ -442,6 +442,7 @@ class LiteLLMParamsBody(BaseModel):
     output_cost_per_token: float | None = None
     extra_headers: dict[str, str] | None = None
     use_in_pass_through: bool | None = None
+    complexity_router_config: dict[str, object] | None = None
 
 
 ModelMode = Literal["batch", "realtime", "image_generation"]

tests/e2e/router/conftest.py

@@ -3,13 +3,80 @@
 The shared lifecycle (resources/scoped_key), proxy liveness skip, and e2e marker
 live in the parent tests/e2e/conftest.py. ComplexityRouterClient holds the shared
 Gateway, so the `resources` fixture cleans up keys this suite creates.
+
+Also registers `complexity-smart-router` via management /model/new when the
+proxy does not already list it (compose has it in static config; stage does not).
 """
 
+from __future__ import annotations
+
+from collections.abc import Iterator
+
 import pytest
+from requests import RequestException
 
 from complexity_router_client import ComplexityRouterClient, build_client
+from e2e_gateway import Gateway
+from e2e_http import NoBody, Success
+from models import LiteLLMParamsBody, ModelsListResponse
+
+ROUTER_MODEL = "complexity-smart-router"
+ROUTER_PARAMS = LiteLLMParamsBody(
+    model="auto_router/complexity_router",
+    complexity_router_config={
+        "classifier_type": "llm",
+        "classifier_llm_config": {"model": "gpt-5.5"},
+        "tiers": {
+            "SIMPLE": "gpt-5.5",
+            "MEDIUM": "claude-haiku-4-5",
+            "COMPLEX": "claude-haiku-4-5",
+            "REASONING": "claude-haiku-4-5",
+        },
+    },
+)
 
 
 @pytest.fixture(scope="session")
 def client() -> ComplexityRouterClient:
     return build_client()
+
+
+def _model_is_servable(gateway: Gateway, model_name: str) -> bool:
+    result = gateway.transport.get(
+        "/v1/models",
+        headers=gateway.transport.master,
+        params=NoBody(),
+        response_type=ModelsListResponse,
+    )
+    return isinstance(result, Success) and any(entry.id == model_name for entry in result.data.data)
+
+
+@pytest.fixture(scope="session", autouse=True)
+def _ensure_complexity_smart_router(  # pyright: ignore[reportUnusedFunction]  # pytest autouse session fixture, wired by name
+    client: ComplexityRouterClient,
+) -> Iterator[None]:
+    """Ensure the complexity router virtual model exists for this session.
+
+    Compose already declares it in docker-compose.yml; stage does not. Register
+    via /model/new when missing and tear down only what we created.
+    """
+    gateway = client.gateway
+    if _model_is_servable(gateway, ROUTER_MODEL):
+        yield
+        return
+
+    try:
+        model_id = gateway.create_model(ROUTER_MODEL, ROUTER_PARAMS)
+    except (AssertionError, RequestException) as exc:
+        if _model_is_servable(gateway, ROUTER_MODEL):
+            yield
+            return
+        raise AssertionError(
+            f"failed to register {ROUTER_MODEL!r} for the complexity router e2e "
+            f"(not listed on /v1/models and /model/new failed): {exc}"
+        ) from exc
+
+    try:
+        yield
+    finally:
+        gateway.delete_model(model_id)