diff --git a/openhands-cli/openhands_cli/tui/settings/settings_screen.py b/openhands-cli/openhands_cli/tui/settings/settings_screen.py
index 983af5a39a..4822555539 100644
--- a/openhands-cli/openhands_cli/tui/settings/settings_screen.py
+++ b/openhands-cli/openhands_cli/tui/settings/settings_screen.py
@@ -174,7 +174,7 @@ class SettingsScreen:
             model=model,
             api_key=api_key,
             base_url=base_url,
-            service_id='agent',
+            usage_id='agent',
             metadata=get_llm_metadata(model_name=model, llm_type='agent'),
         )
 
diff --git a/openhands-cli/tests/settings/test_settings_workflow.py b/openhands-cli/tests/settings/test_settings_workflow.py
index 940a9a802d..192938380e 100644
--- a/openhands-cli/tests/settings/test_settings_workflow.py
+++ b/openhands-cli/tests/settings/test_settings_workflow.py
@@ -18,7 +18,7 @@ def read_json(path: Path) -> dict:
 
 
 def make_screen_with_conversation(model='openai/gpt-4o-mini', api_key='sk-xyz'):
-    llm = LLM(model=model, api_key=SecretStr(api_key), service_id='test-service')
+    llm = LLM(model=model, api_key=SecretStr(api_key), usage_id='test-service')
     # Conversation(agent) signature may vary across versions; adapt if needed:
     from openhands.sdk.agent import Agent
 
@@ -31,7 +31,7 @@ def seed_file(path: Path, model: str = 'openai/gpt-4o-mini', api_key: str = 'sk-
     store = AgentStore()
     store.file_store = LocalFileStore(root=str(path))
     agent = get_default_cli_agent(
-        llm=LLM(model=model, api_key=SecretStr(api_key), service_id='test-service')
+        llm=LLM(model=model, api_key=SecretStr(api_key), usage_id='test-service')
     )
     store.save(agent)
 
diff --git a/openhands-cli/tests/test_conversation_runner.py b/openhands-cli/tests/test_conversation_runner.py
index cdedb3e552..7e39a6f0b0 100644
--- a/openhands-cli/tests/test_conversation_runner.py
+++ b/openhands-cli/tests/test_conversation_runner.py
@@ -50,7 +50,7 @@ class FakeAgent(AgentBase):
 
 @pytest.fixture()
 def agent() -> FakeAgent:
-    llm = LLM(**default_config(), service_id='test-service')
+    llm = LLM(**default_config(), usage_id='test-service')
     return FakeAgent(llm=llm, tools=[])
 
 
diff --git a/openhands-cli/tests/test_directory_separation.py b/openhands-cli/tests/test_directory_separation.py
index 350c3925dc..e95e34aa86 100644
--- a/openhands-cli/tests/test_directory_separation.py
+++ b/openhands-cli/tests/test_directory_separation.py
@@ -37,7 +37,7 @@ class TestToolFix:
         """Test that entire tools list is replaced with default tools when loading agent."""
         # Create a mock agent with different tools and working directories
         mock_agent = Agent(
-            llm=LLM(model='test/model', api_key='test-key', service_id='test-service'),
+            llm=LLM(model='test/model', api_key='test-key', usage_id='test-service'),
             tools=[
                 Tool(name='BashTool'),
                 Tool(name='FileEditorTool'),
diff --git a/openhands-cli/tests/test_mcp_config_validation.py b/openhands-cli/tests/test_mcp_config_validation.py
index 2404e86270..b549768192 100644
--- a/openhands-cli/tests/test_mcp_config_validation.py
+++ b/openhands-cli/tests/test_mcp_config_validation.py
@@ -26,7 +26,7 @@ def _create_agent(mcp_config=None) -> Agent:
     if mcp_config is None:
         mcp_config = {}
     return Agent(
-        llm=LLM(model='test-model', api_key='test-key', service_id='test-service'),
+        llm=LLM(model='test-model', api_key='test-key', usage_id='test-service'),
         tools=[],
         mcp_config=mcp_config,
     )