update

app/core/runner/llm_backend.py

@@ -44,7 +44,10 @@ class LLMBackend:
                 chat_params["stream_options"] = {
                     "include_usage": bool(stream_options_params["include_usage"])
                 }
+        print("stream_optionsstream_optionsstream_optionsstream_optionsstream_options")
+        print(stream_options)
         if stream_options:
+            print(isinstance(stream_options, dict))
             if isinstance(stream_options, dict):
                 if "include_usage" in stream_options:
                     chat_params["stream_options"] = {

app/core/runner/pub_handler.py

@@ -289,6 +289,10 @@ class StreamEventHandler:
             "thread_id": "",
             "metadata": {"usage": chunk.usage.json()},
         }
+        print(
+            "pub_message_usagepub_message_usagepub_message_usagepub_message_usagepub_message_usagepub_message_usagepub_message_usagepub_message_usage==========================================================="
+        )
+        print(data)
         self.pub_event(
             events.ThreadMessageInProgress(
                 data=data, event="thread.message.in_progress"

app/core/runner/thread_runner.py

@@ -139,6 +139,7 @@ class ThreadRunner:
             + tool_call_messages
         )
 
+        logging.info("messages: run %s", run)
         response_stream = llm.run(
             messages=messages,
             model=run.model,