Implement streaming responses for LLM API

app.py

@@ -75,7 +75,11 @@ async def generate_text(request: QuestionRequest):
                 # For more robust streaming, consider Server-Sent Events (SSE) format:
                 # yield f"data: {json.dumps({'token': chunk.content})}\n\n"
                 # For simplicity, we'll just yield the content directly for now.
-                yield chunk['content']
+                if 'response' in chunk:
+                    yield chunk['response']
+                else:
+                    yield str(chunk)
+                
                 await asyncio.sleep(0.01) # Small delay to allow client to process chunks
 
         except Exception as e: