Update app.py

app.py

@@ -268,6 +268,8 @@ generation_config = GenerationConfig(
 pipe = TextGenerationPipeline(
     model=model,
     tokenizer=tokenizer,
+    generation_config = generation_config,
+    task = "text-generation",
     device=model.device.index if torch.cuda.is_available() else -1
 )
 
@@ -344,7 +346,8 @@ async def chat_fn(query: Query):
 @app.post("/ask-rag")
 async def chat_fn(query: Query):
     message = query.message
-    logger.info(f"Received message: {message}")
+    prompt = generate_prompt(message)
+    logger.info(f"Received message: {prompt}")
 
     # Run RAG inference in thread
     loop = asyncio.get_event_loop()