diff --git a/letta/schemas/llm_config.py b/letta/schemas/llm_config.py
index b888a675..27795121 100644
--- a/letta/schemas/llm_config.py
+++ b/letta/schemas/llm_config.py
@@ -164,6 +164,15 @@ class LLMConfig(BaseModel):
                 model_wrapper=None,
                 context_window=128000,
             )
+        elif model_name == "gpt-4.1":
+            return cls(
+                model="gpt-4.1",
+                model_endpoint_type="openai",
+                model_endpoint="https://api.openai.com/v1",
+                model_wrapper=None,
+                context_window=256000,
+                max_tokens=8192,
+            )
         elif model_name == "letta":
             return cls(
                 model="memgpt-openai",
diff --git a/letta/server/server.py b/letta/server/server.py
index 4553de2f..8ee02912 100644
--- a/letta/server/server.py
+++ b/letta/server/server.py
@@ -859,7 +859,7 @@ class SyncServer(Server):
                     value=get_persona_text("voice_memory_persona"),
                 ),
             ],
-            llm_config=main_agent.llm_config,
+            llm_config=LLMConfig.default_config("gpt-4.1"),
             embedding_config=main_agent.embedding_config,
             project_id=main_agent.project_id,
         )
diff --git a/tests/integration_test_voice_agent.py b/tests/integration_test_voice_agent.py
index bc6c09db..b2b8b6d2 100644
--- a/tests/integration_test_voice_agent.py
+++ b/tests/integration_test_voice_agent.py
@@ -215,6 +215,11 @@ def voice_agent(server, actor):
     return main_agent
 
 
+@pytest.fixture
+def group_id(voice_agent):
+    return voice_agent.multi_agent_group.id
+
+
 @pytest.fixture(scope="module")
 def org_id(server):
     org = server.organization_manager.create_default_organization()
@@ -279,8 +284,19 @@ async def test_voice_recall_memory(disable_e2b_api_key, client, voice_agent, mes
 
 @pytest.mark.asyncio
 @pytest.mark.parametrize("endpoint", ["v1/voice-beta"])
-async def test_multiple_messages(disable_e2b_api_key, client, voice_agent, endpoint):
-    """Tests chat completion streaming using the Async OpenAI client."""
+async def test_trigger_summarization(disable_e2b_api_key, client, server, voice_agent, group_id, endpoint, actor):
+    server.group_manager.modify_group(
+        group_id=group_id,
+        group_update=GroupUpdate(
+            manager_config=VoiceSleeptimeManagerUpdate(
+                manager_type=ManagerType.voice_sleeptime,
+                max_message_buffer_length=6,
+                min_message_buffer_length=5,
+            )
+        ),
+        actor=actor,
+    )
+
     request = _get_chat_request("How are you?")
     async_client = AsyncOpenAI(base_url=f"http://localhost:8283/{endpoint}/{voice_agent.id}", max_retries=0)
 
@@ -515,11 +531,6 @@ def _modify(group_id, server, actor, max_val, min_val):
     )
 
 
-@pytest.fixture
-def group_id(voice_agent):
-    return voice_agent.multi_agent_group.id
-
-
 def test_valid_buffer_lengths_above_four(group_id, server, actor):
     # both > 4 and max > min
     updated = _modify(group_id, server, actor, max_val=10, min_val=5)