diff --git a/backend/tests/unit/test_high_priority_usage_tracking.py b/backend/tests/unit/test_high_priority_usage_tracking.py
index 8fa10aaa6a..5c1c2fc644 100644
--- a/backend/tests/unit/test_high_priority_usage_tracking.py
+++ b/backend/tests/unit/test_high_priority_usage_tracking.py
@@ -11,10 +11,16 @@
 import os
 import sys
 import types
+import json
+import importlib
+import importlib.util
+from datetime import timezone
 from pathlib import Path
 from unittest.mock import MagicMock, AsyncMock
 import asyncio
 
+import pytest
+
 os.environ.setdefault(
     "ENCRYPTION_SECRET",
     "omi_ZwB2ZNqB2HHpMK6wStk7sTpavJiPTFg7gXUHnc4tFABPU6pZ2c2DKgehtfgi4RZv",
@@ -28,6 +34,77 @@ def _stub_module(name: str) -> types.ModuleType:
     return sys.modules[name]
 
 
+def _optional_stub_module(name: str) -> tuple[types.ModuleType, bool]:
+    if name in sys.modules:
+        return sys.modules[name], False
+
+    try:
+        spec = importlib.util.find_spec(name)
+    except (ImportError, ValueError):
+        spec = None
+
+    if spec is not None:
+        return importlib.import_module(name), False
+
+    mod = types.ModuleType(name)
+    sys.modules[name] = mod
+    return mod, True
+
+
+# --- Stub minimal LangChain modules needed by usage_tracker ---
+langchain_core_mod, langchain_core_stubbed = _optional_stub_module("langchain_core")
+if langchain_core_stubbed and not hasattr(langchain_core_mod, '__path__'):
+    langchain_core_mod.__path__ = []
+
+callbacks_mod, callbacks_stubbed = _optional_stub_module("langchain_core.callbacks")
+outputs_mod, outputs_stubbed = _optional_stub_module("langchain_core.outputs")
+output_parsers_mod, output_parsers_stubbed = _optional_stub_module("langchain_core.output_parsers")
+prompts_mod, prompts_stubbed = _optional_stub_module("langchain_core.prompts")
+
+
+class BaseCallbackHandler:
+    pass
+
+
+class LLMResult:
+    pass
+
+
+class PydanticOutputParser:
+    def __init__(self, pydantic_object):
+        self.pydantic_object = pydantic_object
+
+    def get_format_instructions(self):
+        return ""
+
+    def parse(self, text):
+        return self.pydantic_object(**json.loads(text))
+
+
+class ChatPromptTemplate:
+    @classmethod
+    def from_messages(cls, _messages):
+        return cls()
+
+    def __or__(self, other):
+        return other
+
+
+if callbacks_stubbed:
+    callbacks_mod.BaseCallbackHandler = BaseCallbackHandler
+if outputs_stubbed:
+    outputs_mod.LLMResult = LLMResult
+if output_parsers_stubbed:
+    output_parsers_mod.PydanticOutputParser = PydanticOutputParser
+if prompts_stubbed:
+    prompts_mod.ChatPromptTemplate = ChatPromptTemplate
+
+pytz_mod, pytz_stubbed = _optional_stub_module("pytz")
+if pytz_stubbed:
+    pytz_mod.UTC = timezone.utc
+    pytz_mod.timezone = MagicMock(return_value=timezone.utc)
+
+
 # --- Stub database package and submodules ---
 # Use _stub_module which only creates if not already loaded
 database_mod = _stub_module("database")
@@ -73,6 +150,7 @@ def _stub_module(name: str) -> types.ModuleType:
 sys.modules["database.knowledge_graph"].upsert_knowledge_edge = MagicMock(return_value={'id': 'e1'})
 sys.modules["database.knowledge_graph"].delete_knowledge_graph = MagicMock()
 sys.modules["database.users"].get_user_profile = MagicMock(return_value={'time_zone': 'UTC'})
+sys.modules["database.users"].get_user_language_preference = MagicMock(return_value='en')
 sys.modules["database.users"].get_people_by_ids = MagicMock(return_value=[])
 sys.modules["database.action_items"].get_action_items = MagicMock(return_value=[])
 sys.modules["database.daily_summaries"].create_daily_summary = MagicMock(return_value="summary-1")
@@ -123,13 +201,45 @@ def _stub_module(name: str) -> types.ModuleType:
 clients_mod.parser = mock_parser
 
 
+def _get_llm_stub(name, **_kwargs):
+    if name in {'goals', 'knowledge_graph', 'daily_summary_simple'}:
+        return mock_llm_mini
+    if name in {'goals_advice', 'notifications'}:
+        return mock_llm_medium
+    if name == 'daily_summary':
+        return mock_llm_medium_experiment
+    raise ValueError(f"Unexpected get_llm feature in usage tracking test: {name}")
+
+
+clients_mod.get_llm = _get_llm_stub
+
+
+def test_get_llm_stub_rejects_unknown_features():
+    with pytest.raises(ValueError, match="Unexpected get_llm feature"):
+        _get_llm_stub("new-feature")
+
+
+def test_optional_stub_module_keeps_loaded_modules():
+    module_name = "test_loaded_optional_langchain_module"
+    loaded_module = types.ModuleType(module_name)
+    loaded_module.Sentinel = object
+    sys.modules[module_name] = loaded_module
+    try:
+        module, stubbed = _optional_stub_module(module_name)
+        assert module is loaded_module
+        assert stubbed is False
+        assert module.Sentinel is object
+    finally:
+        sys.modules.pop(module_name, None)
+
+
 # ── Source-level tests: verify track_usage wraps every LLM call ──
 
 
 def _read_source(relative_path: str) -> str:
     """Read a source file relative to the backend directory."""
     backend_dir = Path(__file__).resolve().parent.parent.parent
-    return (backend_dir / relative_path).read_text()
+    return (backend_dir / relative_path).read_text(encoding="utf-8")
 
 
 class TestGoalsTracking: