[TRTLLM-6842][feat] Support Response API for general purpose (NVIDIA#9392)

JunyiXu-nv · codego7250 · commit 129097dbd24d · 2025-12-12T06:45:10.000Z
Signed-off-by: Junyi Xu &lt;219237550+JunyiXu-nv@users.noreply.github.com&gt;
diff --git a/tensorrt_llm/serve/openai_server.py b/tensorrt_llm/serve/openai_server.py
@@ -994,7 +994,6 @@ async def create_stream_response(generator, request: ResponsesRequest, sampling_
                     use_harmony=self.use_harmony,
                     reasoning_parser=self.llm.args.reasoning_parser,
                     tool_parser=self.tool_parser)
-
         except asyncio.CancelledError:
             if promise is not None:
                 promise.abort()
diff --git a/tensorrt_llm/serve/responses_utils.py b/tensorrt_llm/serve/responses_utils.py
@@ -60,16 +60,13 @@
                                                 StreamingResponsesResponse,
                                                 UCompletionRequest,
                                                 UCompletionResponse)
+
 from tensorrt_llm.serve.tool_parser.base_tool_parser import BaseToolParser
 from tensorrt_llm.serve.tool_parser.core_types import ToolCallItem
 from tensorrt_llm.serve.tool_parser.tool_parser_factory import ToolParserFactory
 
 from .harmony_adapter import HarmonyAdapter, get_harmony_adapter
 
-# yapf: enable
-
-# yapf: enable
-
 REASONING_EFFORT = {
     "high": ReasoningEffort.HIGH,
     "medium": ReasoningEffort.MEDIUM,
