Update app.py

app.py

@@ -5,11 +5,11 @@ import torch
 # Cargar modelo Qwen2.5
 model_name = "Qwen/Qwen2.5-7B"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, device_map="auto")
+model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float32)
 
 # Función de respuesta
 def chat(message):
-    inputs = tokenizer(message, return_tensors="pt").to("cuda")
+    inputs = tokenizer(message, return_tensors="pt")
     output = model.generate(**inputs, max_new_tokens=50)
     return tokenizer.decode(output[0], skip_special_tokens=True)