Update app.py

app.py

@@ -7,7 +7,7 @@ MODEL_NAME = "beyoru/Qwen3-0.9B-A0.6B"
 tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
 model = AutoModelForCausalLM.from_pretrained(
     MODEL_NAME,
-    torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
+    torch_dtype=torch.bfloat16,
     device_map="auto"
 )