move the LLM instance directly to Assistant to make it cleaner to share with tests

src/agent.py

@@ -18,8 +18,6 @@
 
 load_dotenv(".env.local")
 
-AGENT_MODEL = "openai/gpt-5.2-chat-latest"
-
 
 class Assistant(Agent):
     def __init__(self) -> None:
@@ -28,6 +26,17 @@ def __init__(self) -> None:
             You eagerly assist users with their questions by providing information from your extensive knowledge.
             Your responses are concise, to the point, and without any complex formatting or punctuation including emojis, asterisks, or other symbols.
             You are curious, friendly, and have a sense of humor.""",
+            # A Large Language Model (LLM) is your agent's brain, processing user input and generating a response
+            # See all available models at https://docs.livekit.io/agents/models/llm/
+            llm=inference.LLM(model="openai/gpt-5.2-chat-latest"),
+            # To use a realtime model instead of a voice pipeline, replace the LLM
+            # with a RealtimeModel and remove the STT/TTS from the AgentSession
+            # (Note: This is for the OpenAI Realtime API. For other providers, see https://docs.livekit.io/agents/models/realtime/)
+            # 1. Install livekit-agents[openai]
+            # 2. Set OPENAI_API_KEY in .env.local
+            # 3. Add `from livekit.plugins import openai` to the top of this file
+            # 4. Replace the llm argument with:
+            #     llm=openai.realtime.RealtimeModel(voice="marin")
         )
 
     # To add tools, use the @function_tool decorator.
@@ -71,9 +80,6 @@ async def my_agent(ctx: JobContext):
         # Speech-to-text (STT) is your agent's ears, turning the user's speech into text that the LLM can understand
         # See all available models at https://docs.livekit.io/agents/models/stt/
         stt=inference.STT(model="deepgram/nova-3", language="multi"),
-        # A Large Language Model (LLM) is your agent's brain, processing user input and generating a response
-        # See all available models at https://docs.livekit.io/agents/models/llm/
-        llm=inference.LLM(model=AGENT_MODEL),
         # Text-to-speech (TTS) is your agent's voice, turning the LLM's text into speech that the user can hear
         # See all available models as well as voice selections at https://docs.livekit.io/agents/models/tts/
         tts=inference.TTS(
@@ -88,16 +94,6 @@ async def my_agent(ctx: JobContext):
         preemptive_generation=True,
     )
 
-    # To use a realtime model instead of a voice pipeline, use the following session setup instead.
-    # (Note: This is for the OpenAI Realtime API. For other providers, see https://docs.livekit.io/agents/models/realtime/))
-    # 1. Install livekit-agents[openai]
-    # 2. Set OPENAI_API_KEY in .env.local
-    # 3. Add `from livekit.plugins import openai` to the top of this file
-    # 4. Use the following session setup instead of the version above
-    # session = AgentSession(
-    #     llm=openai.realtime.RealtimeModel(voice="marin")
-    # )
-
     # # Add a virtual avatar to the session, if desired
     # # For other providers, see https://docs.livekit.io/agents/models/avatar/
     # avatar = hedra.AvatarSession(

tests/test_agent.py

@@ -1,25 +1,21 @@
 import pytest
 from livekit.agents import AgentSession, inference, llm
 
-from agent import AGENT_MODEL, Assistant
-
-
-def _agent_llm() -> llm.LLM:
-    return inference.LLM(model=AGENT_MODEL)
+from agent import Assistant
 
 
 def _judge_llm() -> llm.LLM:
-    # The judge LLM can be a cheaper model since it only evaluates agent responses
-    return inference.LLM(model="openai/gpt-4.1-mini")
+    # We can use a different LLM to evaluate the agent's responses than the one used in the agent itself
+    # This allows you to use reasoning capabilities or larger models than would be practical for realtime chat
+    return inference.LLM(model="openai/gpt-5.2")
 
 
 @pytest.mark.asyncio
 async def test_offers_assistance() -> None:
     """Evaluation of the agent's friendly nature."""
     async with (
-        _agent_llm() as agent_llm,
         _judge_llm() as judge_llm,
-        AgentSession(llm=agent_llm) as session,
+        AgentSession() as session,
     ):
         await session.start(Assistant())
 
@@ -50,9 +46,8 @@ async def test_offers_assistance() -> None:
 async def test_grounding() -> None:
     """Evaluation of the agent's ability to refuse to answer when it doesn't know something."""
     async with (
-        _agent_llm() as agent_llm,
         _judge_llm() as judge_llm,
-        AgentSession(llm=agent_llm) as session,
+        AgentSession() as session,
     ):
         await session.start(Assistant())
 
@@ -93,9 +88,8 @@ async def test_grounding() -> None:
 async def test_refuses_harmful_request() -> None:
     """Evaluation of the agent's ability to refuse inappropriate or harmful requests."""
     async with (
-        _agent_llm() as agent_llm,
         _judge_llm() as judge_llm,
-        AgentSession(llm=agent_llm) as session,
+        AgentSession() as session,
     ):
         await session.start(Assistant())