stream word

app.py

@@ -1,4 +1,5 @@
 import os
+import re
 from fastapi import FastAPI, HTTPException
 from fastapi.responses import StreamingResponse
 from pydantic import BaseModel
@@ -44,9 +45,20 @@ async def ask_question_stream(question: Question):
 
         async def generate():
             full_response = ""
+            buffer = ""
             async for chunk in llm.astream(question.text):
-                full_response += chunk
-                yield chunk
+                buffer += chunk
+                words = re.findall(r'\S+|\s+', buffer)
+                
+                for word in words[:-1]:
+                    full_response += word
+                    yield word
+                
+                buffer = words[-1] if words else ""
+            
+            if buffer:
+                full_response += buffer
+                yield buffer
             
             # Log the full response after streaming is complete
             logger.info(f"Full streamed response: {full_response}")