Update app.py

app.py

@@ -29,7 +29,7 @@ def say_hello(msg: Annotated[str, Form()]):
     input_ids = tokenizer(msg, return_tensors="pt").to("cpu")
     print("output")
     inputs = tokenizer.encode(input_text, return_tensors="pt").to(device)
-    outputs = model.generate(inputs, max_new_tokens=256, temperature=0.6, top_p=0.92, do_sample=True)
+    outputs = model.generate(inputs, max_new_tokens=32, temperature=0.6, top_p=0.92, do_sample=True)
 
     print("complete")
     return {"message": tokenizer.decode(outputs[0])}