Update app.py

app.py

@@ -16,7 +16,11 @@ model_name = "meta-llama/Llama-3.2-1B-Instruct"
 device = 'cuda' if torch.cuda.is_available() else 'cpu'
 
 model = AutoModelForCausalLM.from_pretrained(
-    model_name, torch_dtype=torch.float32, device_map=device
+    model_name,
+    device_map="auto",
+    torch_dtype=torch.bfloat16,
+    attn_implementation='eager',
+    trust_remote_code=True,
 )
 
 checkpoint = "gregorlied/Llama-3.2-1B-Instruct-Medical-Report-Summarization"