revert print

llama_stack/providers/inline/agents/meta_reference/agent_instance.py

@@ -168,17 +168,10 @@ class ChatAgent(ShieldRunnerMixin):
             if self.agent_config.instructions != "":
                 messages.append(SystemMessage(content=self.agent_config.instructions))
 
-            from rich.pretty import pprint
-
-            print("create_and_execute_turn")
-            pprint(request)
-
             for i, turn in enumerate(turns):
                 messages.extend(self.turn_to_messages(turn))
 
             messages.extend(request.messages)
-            print("create_and_execute_turn turn to messages")
-            pprint(messages)
 
             turn_id = str(uuid.uuid4())
             span.set_attribute("turn_id", turn_id)
@@ -367,7 +360,6 @@ class ChatAgent(ShieldRunnerMixin):
         documents: Optional[List[Document]] = None,
         toolgroups_for_turn: Optional[List[AgentToolGroup]] = None,
     ) -> AsyncGenerator:
-        print("_run messages", input_messages)
         # TODO: simplify all of this code, it can be simpler
         toolgroup_args = {}
         toolgroups = set()
@@ -498,7 +490,6 @@ class ChatAgent(ShieldRunnerMixin):
             stop_reason = None
 
             with tracing.span("inference") as span:
-                print("just before chat completion", input_messages)
                 async for chunk in await self.inference_api.chat_completion(
                     self.agent_config.model,
                     input_messages,

llama_stack/providers/remote/inference/together/together.py

@@ -196,8 +196,6 @@ class TogetherInferenceAdapter(ModelRegistryHelper, Inference, NeedsRequestProvi
         logprobs: Optional[LogProbConfig] = None,
     ) -> AsyncGenerator:
         model = await self.model_store.get_model(model_id)
-        print("inside together chat completion messages", messages)
-        breakpoint()
         request = ChatCompletionRequest(
             model=model.provider_resource_id,
             messages=messages,
@@ -225,11 +223,6 @@ class TogetherInferenceAdapter(ModelRegistryHelper, Inference, NeedsRequestProvi
 
     async def _stream_chat_completion(self, request: ChatCompletionRequest) -> AsyncGenerator:
         params = await self._get_params(request)
-        from rich.pretty import pprint
-
-        print("together stream completion")
-        pprint(request)
-        pprint(params)
 
         # if we shift to TogetherAsyncClient, we won't need this wrapper
         async def _to_async_generator():
@@ -247,7 +240,6 @@ class TogetherInferenceAdapter(ModelRegistryHelper, Inference, NeedsRequestProvi
     async def _get_params(self, request: Union[ChatCompletionRequest, CompletionRequest]) -> dict:
         input_dict = {}
         media_present = request_has_media(request)
-        breakpoint()
         if isinstance(request, ChatCompletionRequest):
             if media_present:
                 input_dict["messages"] = [await convert_message_to_openai_dict(m) for m in request.messages]