improve

llama_stack/providers/remote/inference/tgi/config.py

@@ -15,10 +15,6 @@ class TGIImplConfig(BaseModel):
     url: str = Field(
         description="The URL for the TGI serving endpoint",
     )
-    # api_token: Optional[SecretStr] = Field(
-    #     default=None,
-    #     description="A bearer token if your TGI endpoint is protected.",
-    # )
 
     @classmethod
     def sample_run_config(cls, url: str = "${env.TGI_URL}", **kwargs):

tests/client-sdk/inference/test_inference.py

@@ -225,7 +225,6 @@ def test_text_chat_completion_with_tool_calling_and_non_streaming(
         tool_prompt_format=provider_tool_format,
         stream=False,
     )
-    print(response)
     # No content is returned for the system message since we expect the
     # response to be a tool call
     assert response.completion_message.content == ""