Update app.py

app.py

@@ -72,8 +72,8 @@ device = "cuda:0"
 # Load model and tokenizer outside the generation function (load once)
 print("Loading model...")
 model = AutoModelForCausalLM.from_pretrained(
-    "m-a-p/YuE-s1-7B-anneal-en-icl", # "m-a-p/YuE-s1-7B-anneal-en-cot", 
-    torch_dtype=torch.float16,
+    "m-a-p/YuE-s1-7B-anneal-en-cot", # "m-a-p/YuE-s1-7B-anneal-en-icl",
+    torch_dtype=torch.bfloat16,
     attn_implementation="flash_attention_2",
 ).to(device)
 model.eval()
@@ -358,7 +358,7 @@ with gr.Blocks() as demo:
         gr.Examples(
             examples=[
                 [
-                    "rap piano street tough piercing vocal hip-hop synthesizer clear but aggressive vocal male",
+                    "rap piano street tough piercing vocal hip-hop synthesizer clear vocal male",
                     """[verse]
 Woke up in the morning, sun is shining bright
 Chasing all my dreams, gotta get my mind right