Update app.py

app.py

@@ -73,7 +73,7 @@ device = "cuda:0"
 print("Loading model...")
 model = AutoModelForCausalLM.from_pretrained(
     "m-a-p/YuE-s1-7B-anneal-en-cot", # "m-a-p/YuE-s1-7B-anneal-en-icl",
-    torch_dtype=torch.bfloat16,
+    torch_dtype=torch.float16,
     attn_implementation="flash_attention_2",
 ).to(device)
 model.eval()