Updated app

app.py

@@ -79,7 +79,7 @@ def generate(req: PromptRequest):
 
     try:
         # Limit tokens to avoid huge outputs
-        max_tokens = 200
+        max_tokens = 150
 
         output = pipe(req.prompt, max_new_tokens=max_tokens, do_sample=True)
         text = output[0].get("generated_text", "").strip()