update oai

qwen_agent/llm/qwen_oai.py

@@ -326,8 +326,7 @@ def create_chat_completion(request: ChatCompletionRequest):
         if request.functions:
             raise Exception("Invalid request: Function calling is not yet implemented for stream mode.")
         generate = predict(query, history, request.model, stop_words, gen_kwargs)
-        print(generate)
-        return json.loads(generate)
+        return generate
         # return EventSourceResponse(generate, media_type="text/event-stream")
 
     stop_words_ids = [tokenizer.encode(s) for s in stop_words] if stop_words else None
@@ -506,6 +505,7 @@ class QwenChatAsOAI1(BaseChatModel):
                                                 stream=True)
         # TODO: error handling
         for chunk in response:
+            chunk = json.loads(chunk)
             if hasattr(chunk.choices[0].delta, 'content'):
                 yield chunk.choices[0].delta.content